Making of a method

At this point, I ask that those who have questions of me and/or of what I do to post them in my journal. This is Game's journal, not mine.
 
Quote from dbphoenix:

As regards this left to right business, I should emphasize to those who are pursuing this type of trading that when conducting this sort of (discretionary subjective blah blah) backtesting, it is extremely important that one not allow whatever foreknowledge one has to influence his trading decisions. Here, for example, it is a grave error to pass up the bottoming action on the 5th because of the upcoming retracements. In real time, one cannot know there will be two retracements. Or even one. Or any at all. In real time, all one has is the bottoming action. Therefore, he has to evaluate that action as it happens and make the proper risk assessment. If he decides that the risk is not worth it, at least to him, then he is entitled to pass. However, he should not then beat himself up for having passed in case there is no subsequent retracement. Just as important, he should not pat himself on the back for having passed on the bottoming action just because a subsequent retracement presented him with another entry opportunity and try to apply that happenstance to the next situation which appears to bear some similarity to this one. Again, the bottoming action has to be evaluated as is. Otherwise, all the time and effort spent backtesting is wasted. [/QUOTE


How does prep work fit into backtesting? I am assuming that the general tone and important past S/R levels are not a part of backtesting, since the trader is just going bar by bar and looking for certain patterns that reflect the hypothesis.

If my hypothesis is based on taking Reversal trades off S/R after a re-test (where the re-test is defined specifically as x,y,z), am I then obliged to include it in the results everytime that it shows up during the backtest?

In Real Time, the presence of a very strong and anticipated past S/R level may prompt me to take an aggressive trade without waiting for too much of a re-test. Or something else may prevent me for taking the trade.

So do I ignore these nuances for now and just limit the variables to patterns that can be backtested bar by bar?
More...
 
Quote from game:

How does prep work fit into backtesting? I am assuming that the general tone and important past S/R levels are not a part of backtesting, since the trader is just going bar by bar and looking for certain patterns that reflect the hypothesis.

If my hypothesis is based on taking Reversal trades off S/R after a re-test (where the re-test is defined specifically as x,y,z), am I then obliged to include it in the results everytime that it shows up during the backtest?

In Real Time, the presence of a very strong and anticipated past S/R level may prompt me to take an aggressive trade without waiting for too much of a re-test. Or something else may prevent me for taking the trade.

So do I ignore these nuances for now and just limit the variables to patterns that can be backtested bar by bar?
More...

Depends on your hypothesis. But the world does not begin with the left edge of the chart. Trend and S/R must be taken into account with nearly any hypothesis (I say "nearly" though I can't think of any offhand that wouldn't apply).

S&R don't come from nowhere. And one must always take the overall trend into account. And, of course, price doesn't move bar by bar.

If you can provide specific charts and specific examples, I can be more specific myself. If you do, however, try not to stretch them out so much. The action is much easier to see in something closer to a 1:2 ratio.
More...
 
Quote from dbphoenix:

You appear to be approaching an understanding of what this is all about, not about the price per se, but about the behaviors -- the desires, the wants, the anxieties and so forth -- that bring about the transaction that results in the print in the first place.

Good on you.
More...

The words fear and hope are often used to explain the underlying basis behind trading behavior. However, in light of the progress made by developing frameworks to explain this behavior via Prospect Theory, etc, are you aware of any work that uses principles like loss aversion in a systematic way to explain behavior? Something akin to what Mamis did with the sentiment cycle in 'How to Buy'.

If boundary conditions offer the best bets, are there predictable emotional markers that can be mapped at these points?
 
Quote from game:

The words fear and hope are often used to explain the underlying basis behind trading behavior. However, in light of the progress made by developing frameworks to explain this behavior via Prospect Theory, etc, are you aware of any work that uses principles like loss aversion in a systematic way to explain behavior? Something akin to what Mamis did with the sentiment cycle in 'How to Buy'.

If boundary conditions offer the best bets, are there predictable emotional markers that can be mapped at these points?
More...

Not familiar with Prospect Theory, though I know what it is. I try to avoid thinking about this stuff too much. It clouds judgement. The idea of loss "aversion" doesn't help much since people go through all sorts of mental and emotional gymnastics to avoid acknowledging loss (e.g., "it's not a loser until you sell it"). So I'm not as concerned with the psychological problems of traders as I am with what they do.

As to predictable emotional markers, no, not predictable. Revealing, perhaps. Such as the way ranges become tighter and tighter within a given timeframe as price reaches the crest or trough of a move before reversing or busting out. Or the little hesitations that occur after a breakout that suggest temporary confusion and give the alert trader an opportunity to enter a trade that he missed. Most traders are confused and doubtful and anxious. If you're not and if you can recognize these behaviors in others, then you're ahead of the game while everybody else is standing around wondering what just happened.

This, by the way, is my primary quibble* with people who write books on price action and focus on bars and what this bar means or what that bar means. The bars don't "mean" anything. It's the behavior of the traders who created the bar in the first place that matters. If one spends his time focusing on what the bar means, he's going to spend so much time figuring it out -- or trying to -- that he'll miss the trade entirely.

*My secondary quibble has to do with those who say they don't use indicators then plot moving averages and use colored candles. (?)
 
http://www.sierrachart.com/image.php?l=1374535334495.png

Prep:
Price at lower boundary of Daily channel. After spending most of Friday at the 3031 level, price turned up to close on strength. ON upward drift found R and then a retest was successful through a LH before PM. Price has been making LH's since then, but there was a 6 point surge at 0828. No strong bias for now.

S: 3041
S: 3031

R: 3048
R: 3052

0830 Very strong buying pressure taking price beyond immediate R to 54.
0832 price dips below mp at 48. Brought back strong to close above mp. A lot of uncertainty at this point.
0836 Price climbing. Likely to test R
0837 Retest attempt may have made LH and this may be the beginning of the downswing as price just crossed DS line.
0844 If short, no reason to exit. Can see what happens at 41 level.
0850 No strong buying pressure. The reversal attempts could not go past the SH level.
0852 selling pressure continues. LL. the ds line rotated out. But most likely sideways action since selling pressure is not too strong either.
0856 Break through S level of 41 and traders could take it back to yesterday's 31 zone. Short bias on.
0900 The buying surge indicates that this may be the bottom for now. Nothing to do for now unless price leaves this congestion.
0905 Could be considered as a retest of S. But seems to be a low quality situation considering that the re-test is occurring within the context of sideways movement. So my rules will need to reflect the presence of waves in order for the retest marker to be valid for a trade. Otherwise it is just a breakout situation out of this congestion zone.
0910 the upward drift leads to a breach beyond SH point. mp is 45.50
0913 Price finds R at mp and goes back down. Don't feel that this current environment offers any compelling opps. Pace has slowed and buying pressure has been inching up. Lot of trades below at S and we have already had two big waves for the day. So not much movement likely now.


1) The compelling opp today because
a) The continuation move was weak as shown by the buying waves getting smaller
b) There was a retest of R.
c) Short was just below anticipated R level of 52 that stopped the continuation attempt.

2) Not as compelling because
a) Taken in the sideways drift action after the completion of a big wave


Ideas:
Rule for retest: The first reversal wave has to show some strength via extent and pace. The retracement of this wave will be considered the retest attempt. Otherwise there will always be multiple re-tests in close proximity within a chop situation that becomes a small range.
 
Although 1 was a very good entry in hindsight, what where your reasons to take it in Real Time, besides price hitting Resistance I could not find any other thing that allowed me to take that short, I kwow I am the one who is wrong, or else I would have taken that short :(.
 
Quote from niko:

Although 1 was a very good entry in hindsight, what where your reasons to take it in Real Time, besides price hitting Resistance I could not find any other thing that allowed me to take that short, I kwow I am the one who is wrong, or else I would have taken that short :(.
More...

It was not an entry, just an observation of a potential entry. I am not sim trading till I have a backtested plan. My thinking at the time was:

0832 price dips below mp at 48. Brought back strong to close above mp. A lot of uncertainty at this point.
0836 Price climbing. Likely to test R
0837 Retest attempt may have made LH and this may be the beginning of the downswing as price just crossed DS line.

I have been focusing on re-tests as my current hypothesis is that a retest scenario offers the lowest risk through it's combination of both lower price and information risk. Still trying to define and qualify the various markers involved with the set up, so I don't have all the details yet.

The quality of the buying waves in the continuation attempt was indicating upcoming weakness. Once a re-test confirmed R, the short could have been taken on the break of DS line. There was another hesitation 2 mins later as well.
 
It's time to define the set up to begin backtesting. The goal not to be either too inclusive or too exclusive when defining the markers. Task ahead is to make it simple and robust. Goal is to keep the bigger picture of risk in mind and whether the markers are reflecting the hypothesis. I can always tweak in subsequent runs.

Some elements of the Reversal set up

Quality of S/R - Past importance. Extent, time and pace of first rejection.

Quality of Reversal wave - Extent, time and pace. Does it bust through? Waves getting smaller? Hesitations everywhere?

Quality of S/R re-test - Type of re-test via HL,DB, LL,. Lot of sideways movement in this zone or touch and go?

Quality of the hesitation preceding the stop

Veto Conditions - waves within LH and HL, other congestion/range type behavior.
 
You must log in or register to reply here.
Back
Top