Pattern's Chemo-Electric Trading Journal

Returned from 'bar channels' thread. Was there laying some groundwork for the trendlines I use, which are based on bar spreads. Left because I lost my way trying to make spread lines a method; that's why I'm here this time.

Okay, so I gotta make spread lines a method. I'm fairly sure it'll work, because I was getting stopping lines along a run, small lines that showed a turn, and generalizing lines that the price bounced around for a good while more or less.

Oh, and just to be pc, I lost 4-5 accounts past several days, because I'm undermining my instinctual-trained ability to trade, to represent them instead.

Brought in news, after several recent threads, each with highly competent posters, were posted. Easy enough: Avoided time happens, returned to read effect news had, and applied pattern-line table to it. All I'm doing is using the minute the any ff red-flag news happened in to apply some sort of plt study to. For example, my first study was of 10Jun09 1330GMT on g/u; the 1'-condensed trend surrounded the bar exactly, so I went for a centering idea that's temporary. 89sma centerline fit the bill, only the 89sma didn't relate well, so I took the midpoint of the 1330 minute bar and drew a ray from it. This study paid off well, hit several major trends/turns. Never would've known it close-up or with the immediate pattern the news bar was in.
Did another news bar. This one was even more interesting. It was 1330GMT 06112009 g/u. It fed off yesterday's news bar. No kidding! Even more, yesterday's news bar pivoted the 1'-chart trend 0611's bar was in, and there was an 89sma centerline that exactly -- yes, exactly -- went between them. AND, when I put deviations on the three of them, they finished projecting the rest of the trend's levels, plus the top consolidation and the small drop after it. Bar channels, crosschannels, balance lines and low-spread areas just showed the smaller jumps as the price followed the news/89sma centerlines.
So I'm wondering, it's worth checking to see how involved news and 89sma are with each other. At least, 89sma centerlines became a lot more important.
http://i42.tinypic.com/15xoe39.jpg http://i41.tinypic.com/alq82h.jpg http://i43.tinypic.com/2uz3ti1.jpg
Also, the 0930GMT06112009 g/u newsbar centerline did work with the 1330 of 10 June, only it stopped being useful once next day's pre-(89sma c.),1330-c.'s hit. http://i40.tinypic.com/2rh4ojc.jpg http://i39.tinypic.com/2cy3yfs.jpg
I'm sure news, even at this level of simplicity, is far more complicated than the above. There were only these three ff's-red-flag news bars for the entire week, which simplified everything. I looked at the week before, and whoa, I gotta lotta work to do before this idea works regularly and not only with exceptions.

Here's another one, more along the idea of how I used to trade news, a bit in forex and 10y in stocks: http://i44.tinypic.com/2po9itl.jpg . Going thru the study should help explain a lot of those big jumps, regardless of prt also doing so.
Alright, I can't stop. 060209 1500's big surprise centerlined really well for the top of the next week's later-in-the-week's run, and 06042009 0900's big surprise centerlined for the bottom of the next week's later-in-the-week's run. So big-difference news bars into an effective centerline.
Also in both of these surprises were charts that anchor the next week's moves. They also look unusually cohesive in approach. Check the latter's: http://i43.tinypic.com/330a5wp.jpg .


-----

demo trades: A little confidence on a few trades, just line dependency right now, working to bring in prt (which am now calling pattern-line areas, or something..)

-----

Okay, so basically newsbar centerlines work. They focus price until another big news surprise occurs.
I fiddled with them and also found pattern copying related to them: http://i41.tinypic.com/qqs7li.jpg , http://i41.tinypic.com/2daycmx.jpg . Given some of news surprises' dramatic and different patterns, it's worth studying. I've seen this effect while looking at Time Zones vertical lines drawing tool, just didn't have a reason to anchor on.

(continued from last post)
Copied patterns from newsbar centerlines'd increase the method's precision significantly. I did find some more copying.
Moreover, I found another principle to newsbar centerlines. Of course; trouble has to show, ha-ha. June5 1330 newsbar of NFP surprise anchors with midlines prices thru the entire following week, irrespective of the other significant values (June 4 0900, June 2 1500).

So I have newsbar centerlines centering significant amounts of prices, at times copying patterns around them, and at times anchoring with midlines prices around them.

-----

Oh, to be with it, no trading yet, platform's not listing prices yet for Sunday open.

-----

So, even though haven't tested these extra features of newsbar centerlines, want to rally the more important bar channels and 89sma centerlines around them to keep efficient. I say this, because I accidentally on purpose kept an important trendline on board, the ~1130 June 4 value of 104 on CDRW Spread that channels with the ~1245 June 5 value of 55 on CDRW Spread.
Apparently this bar channel is also news-related. That's probably a fourth feature of newsbar centerlines. Guess I'd better check on June 2, which'll take a while. (Unless I can get away with not doing it...)

-----

Alright, trading started, only with me doing all this very basic and large-area information, I don't really trust me to trade beyond instinct..

Found a couple of lines, got a few pips... At least began... (2015EST June15)

----

(1144EST) On news: I think I need to know what's a good number and what's a bad number for each news item, while avoiding the slippery slope of actually reading the and into the news' impact. I was thinking of connecting all of the high-impact newsbars into one line; knowing good and bad direction for each variable and then comparing this to a newsbar line would clarify, to a close-enough degree I believe, what variables (news items) would be more or less important. Intensity of a high or low miss would, I hope, be reduced as being needed to know, beyond the initial red-flag obvious big miss screen for a news item to make it to being a newsbar.

Lost several accounts.

-----

Figured news.
You draw a newsbar centerline for the overall level perspective (tested April 8 to current date for all extreme-miss news items (excluding qualitatives like speeches) and a crosschannel along the declining or ascending values, depending on the gradual selloff or buy build and thus having on the right or left the extreme-low CDRW Spread value being the balance bar and on the left or right the locally-high CDRW Spread value which is traditionally drawn according to bar channel instructions.

prt gonna work, what with the new newslines? Got a helper in there (a 50-60), still....

Return to the drawing board...


----

trading: Actually won a few trades, then off the rocker with the above.
At least, the lines are right. ( )

I just realized, that's a funky chart above. I was grouping patterns into nice, neat areas that happened one right after another, when really what these areas are doing is playing off each other. I never considered this in my prt use, and whether or not it changes prt, which I doubt, or changes my use of it, which I'd guess is correct, I need to figure.

Alright, here's a bit of figuring:



The bright blue lines figure the lead-in trend and then the trend, with double top responding to newsbar centerlines in yellow, that double-bottoms. The red lines, to the high spread that's after it, work the center between the double tops at the highlighted CDRW Spread formation.
So what this highlighting of certain lines (minus the newsbar centerlines and other lesser differentiation of each basic trend) tells me is that
there's one downtrend leading in, with the first bright blue line figuring its turn;
next, there's an uptrend, which stays high for a while bouncing amongst the newsbar centerlines;
then, an downtrend leads from the middle of all that bouncing and to a lower newsbar centerline, with the 50-60 white channel bringing it short of touching that n.c. (even though it obeys the bright-blue pair from the double-top's middle more);
and finally, a proposed uptrend within the red pair of lines that megaphones from the middle of all that bouncing between the upper n.c.'s.

(Please, anyone reading this, I just got lucky. There is a whole lot of implied causation to the above list, only it's my first realization of this real order of things.)


-----

(7:09 pm EST) Hey, it worked. Now, I don't know exactly why it worked this well (though, as noted, there's accrued a decent number of causative connections to choose from.) I'm working on it, hoping (yeah, right) prt will pull thru.



-----

(5:10 am EST) Need more proof?



The lines work.

They work!! Too bad the patterns don't...



WAIT. I switched platforms and something messed up on the lines, including the red lines pivotal to my claim that the lines work.
I'll have to review what I've done, repeat the process several times, and include patterns.
I did glance at the picture above, and patterns did work near the line as if the line was the proper line, where PA hits the red line.
I just need at least the weekend to get my act together on this.

-----

Also, want to break down news lines a bit more based on finding at least the most exceptional of all the exceptional news items to n.c. on. Possibly, I may bring in midrange flat spread areas, which may lead to a cleaner understanding of why spread-area deviations work so well.

-----

Didn't really trade today. Trying an ECN mt4 platform.

I keep running into good ideas. Just today, I read forex has a large following in its options, so options expiry timing is a good indicator. Another is the widely-publicized tone of the US stock market opening and its first 15-75 minutes. (For that matter, the first week of earnings season, when the biggest corporations report, is also notable in the 15 minutes after NYSE close and next day's open.) A third is the price which has the highest futures volume for a certain period. A fourth is major speeches that have a really strong reaction in EUR/USD, which is easier to read, to me, at these times for such a purpose. Down the road, I'll have time to bring these in.

method description:
Prt responds best to lines and the occasional distinctive pattern. Apply spread trendlines to spread-trend highs/lows (where lines are drawn at low points between, mid-range areas within or between, and high points breaking up spread trends), which help guide PA's pressure; newsbars (where newsbar centerlines, each newsbar being the minute of announcement of a quantitative forex news event which prediction was missed severely), and their groupings are studied), which center the price in the bigger picture for a while; balance lines of 2-3 high spread bars, which project the (or one of the) next trend(s) ; deviation lines, which may arrive from any line, including ordinary, temporary spread trendlines that pitch in and, placed right according to the guide - stressor - trend - countertrend - balancer labelling approach, to cover ideas missed such as the last paragraph's that the prt will respond to; and, pattern recognition table readings of lines' effect on inter-, copied patterns of PA's interaction with these lines, both how different types of patterns react to each other to make an area and how features of these patterns change when repeated next in the same or similar area.


The picture below shows spread trends ("deviations") (red trendlines) and low points between (green frames). The idea is similar to the consol and trend idea in the 'EUR/USD Patterns' thread.

News may further subdivide into several highest-impact news and the rest red-flag news. A first guess would be NFP, FOMC, ... US Fed Rate and a few others may be worth listing even without a surprise number. (Color, gold.)

-----

Trading is a little better, and laying forth a new version of my method helped a lot. I know how far I have to go to produce a good or better trade within the method, at the same time that I have definite limits as to how much I [i]ever[i] have to do to get a good trade going. That's very comforting.

My first trade made +55% on my demo account, Sunday open, in fifteen minutes.
Reason, two newsbar centerlines from Friday based a run each, falling between. Today is Sunday: No news. Thus, any news'd be digested, and Friday news'd be discounted; plus, there was a big runup on Friday, from spread-trend crosschannels across the week. Ah, heck, I'm guessing on the news. Really, it was the pattern repetition that got me from one newsbar centerline to the next. This was enough to get me in at the right place, only I missed the exit by a landslide: The n.c. also. I felt I would have had to apply the table to move myself to consider an exact exit.

-----

On platform times mixup: More develops. I mismatched my Forex Factory time to my platform's time. I had done GMT + daylight savings time, instead of the standard GMT only, on ff. The OD.L / 'MetaTrader' platform is always timed to GMT, so I'm using it now; I also use Insta.Forex, which is GMT +2.
There is no way to change a platform's time, I checked metatrader's site today. Also, you can't move a template full of drawing tool objects from a platform with one time to that of another time. I'll have to redo my first posts to reflect this.

-----

Well, how about that. Third news item I recheck due to the above mixup, and it's a doozy. 63 spread. Wow.
I'm glad I did make the mistake, though; else, I may not have thought up newsbar centerlines. Again though, reworking the chart, they may not work now. I do hope I can narrow the number of lines to use.

-----

(820pm EST)Another trade. Pulled it short. Reason, a bar channel that worked decently late last week to turn PA, and a basing of PA in an overall N pattern ('harami', to me). Why pulled it short: Was erasing my old, inaccurate newsbar centerlines (that somehow worked in some cases -- Odd.) Caused insecurity, which compounded the usual decay cycle of my confidence after a good, clean trade.

-----

Tell you what. I'll leave those lines that 'don't' work on one platform, and check with them later, once I get the real ones up on another.

------

Took a too-wide reason, margin call.

-----

(morning: 0524am EST Jn 22 9) Lost another account. Again, I mismatched my reason accuracy to my trade margin; that is, too wide of a reason. Like the last one, this one also traded fine, otherwise. This is the only trading skill I probably need to learn, so I'll obtain and use it.
It seems the mismatch occurs relating entry to exit. I"m sure the move will allow a profitable exit, and this is the second thing I think after I check with the entry. I'm treating the entry as the entire move being okayed, which is leaning the trade toward being theoretical or a cheaply-done one. Anything done poorly, gets the same results; invest a lot on something poor, and drastic accidents occur.
So I'm making a huge mistake, here, and not being honest with myself. My reasons stink occasionally. I have to go with the flow of how good they are, in order to perform this trading skill of matching reason accuracy to trading margin.

-----

Did it again. This time, fuzzy on reason, trade fell apart immediately. (1115amEST)

-----

Spread-trend lines, which the most-closely correlate with traditional pivot-based trendlines, form-fit to PA the most. They do, however, work really, really well, and are essential to my method.

-----

money management:
Realized something very important. Stops have an innate place to exist in. Often, I want a reason to work, then the price goes directly against my reason. That reason was a stop reason.

Lost account several times. Last time, thot the above. (716pm EST)

technical analysis:
I was able to get decay features into my lines, thru deviations. What I mean is, a line, thru PA's relation to its deviations, works some then peaks in its working then fades and finally stops in its working, that is, has a decay cycle of working for a particular area of PA.
This feature supercedes the guide - stressor - trend - countertrend - balancer organizational approach, or at least works with it.

-----

current trade, minus longterm reason:
(green lines only) (orange lines only) (green, orange)

Green lines based on an ordinary 22 spread and the first almost-0 spread with a distinctively-low area with it.
Orange lines based on an unusual occurence of four 1-pip-wide bars.
personal note: I had drawn the lines midway in the uptrend shown before PA hit the orange line.

The longterm reason was on a 55-62 bar channel that crossed rather early, did well on deviations for alot of the day, and seems now to have PA head for its main lines far below. Its deviations, thick yellow trendlines (shown).

money management:
I lost several accounts before realizing, by chance and with some resistance to the notion before this, the longterm crosschannel (not shown) would work. Once I did, I gained confidence in using local lines. I took a longterm, small position to hold, and added temporary, larger positions with shorter bounces.


-----

(0540pmEST)Overnighter trade, a short-term high-margin trade, worked. My account's up 67% and I can, after the above-noted notion, justify it completely. I believe I made a breakthrough in my trading career. This reminds me of my favorite moments in the past 12 1/2 years: KTEL, the $11,000 profit, the Week, the Beazer Homes technical trade, and the Chat and Spin demo trade. Really, all of these pale in comparison.(Of course, the 1000% in 17 hours demo trading doesn't.)

-----

(0354pmEST) !#$%$@#%^&^$!!!!!!!!!!!! Lost account! Totally ignored all lines. You know what really pisses me off? I should have. I lack trust of my with-trend lines, still, and no reason should stop me from doing so. Apparently, that's my next task.

-----

Alright, the drawing board.
trendlines.
aarghhhhhhhhhhhhhhhh...........................

-----

One thing in trading... Something that should happen, never will. You must have concrete reasons to say it'll do so, and understand there are often reasons you'll miss, due to doing the right thing only not enough, doing the wrong thing and too much. It's always about the right reasons, both overview and details, area and trend.
Alright, area and trend. Actually, condensed-chart area-trend, in which you have to go in to trade anything, and just-touching-chart in-area trend, in which you have to go wide to trust anything. The two interweave, so you have to keep up with it

-----

Okay, I guess I got a grip now.

It just surprised me that some seat-of-the-pants line I stuck on the chart is now the star of the show. Plus, a big 55-62 deviation isn't knocking down the price into the main line for that 55-62, what with FOMC on the horizon in a couple of days. Facts, dontcha hate 'em.

I'll run a bunch of examples and see if I can get this portion going as trustable/dependable.

(340amEST) I lost my account because I was bull-headed. My idea went thru fine. I didn't place my margin according to what reasons set I had to trade with. Who am I trying to kid with too big a margin, in this case?

Another day.

new account, trade: I mistrust lines. That's okay, the stream of logic was poor. That's not okay, because the face value was high: http://i42.tinypic.com/97v709.jpg . PA was just using the old lines again, and the old lines were well-chosen, I did have a good session when I made the lines, everything was causative.
So I got this first trade, only with 2/3 maximum drawdown because of the mistrust.

lines mistrust for margin choosing per each PA point --
learning verbatim from a successful session vs. imitating, pragmatizing after it
old lines: Area performed well in, stream of logic/causation ; more specifically, new lines worked up for new overall direction, using older lines for levels or both in 45d angle of new and old
surface appearance of lines currently
line worth, set-of-lines reason, for current into future
impulsivity in seeing a small reward now or big one later

I watched Nova ("Genius Apes") on PBS last night, and one of the things they were talking about, in one experiment about a treat being right in front of them and the overseer offering a bigger treat later on, was the game of wait and get a lot or take now and get a little, which judges impulsivity/patience.
In a second experiment, they talked about how apes imitate, that is, learn only enough until their senses tell them the reward is easier to obtain than the copied action suggests. This is versus people, who have the capacity to learn, that is, even in the face of an easier way to get the reward, will delay gratification in this instance in order to follow instructions exactly.
While these few experiments have a little more to do with overseer-worker relations than trading does, they suggest how we may act in trading. I acted very impulsively when I put too high a margin on my trade that lost my last account. This compounded the basic fault that my reason to trade was worth very little, actually was worthless, where I chose to enter my trade. Yet, because of the first experiment's result, I jumped at the chance, impulsively. The last trade, after the lost account, was an even more obvious case of impulsivity, based only on a thin veneer of mistrust of lines whose face value was high.
Because of this, I added the last two statements to the above pre-trade checklist.

A problem with using old trendlines on a reverse is that they are flatter and are more checkpoints in a trend only newer lines can interpret, so you don't get out of the work. (Sad/)

(0900am EST) Changed the order of rules, making step 5 step 1, since it implies how good one's method is in the first place. The order shows a simple refocusing from wide to narrow, in terms of trade.

redoing success verbatim (boring) vs pragmatize (fun) vs both (difficult, boring, like eating your greens):
Both would be to break down some of the learned material, that which looks to fail now, and then to pragmatize a replacement for it and test it until it's worth learning verbatim. Sometimes you have to tear down the entire body of work, other times just a little will do. During trading, impulsivity vs. spontaneity of reason to trade will tell one where in this range to go, and decay of interest vs. searching elsewhere for an answer how often to repeat the process.

-- rewrote the list in the last post


Smoothly deducing pattern shape from a verbatim learning of lines use from two nights ago, slowly letting this new understanding take over the more-difficult work of finding and setting up those lines, would be optimal.

-----

Trades, abysmal.

-----

Just had a thought. (Ouch.) If I do the opposite, I may get better. Usually, I approach the MT4 trading frame with fear and impulsivity, and trade too big and too often. Why? I may have a good motivation. There is the tick chart plus the moving price per pip worth; these inform me of the most vital of the chart's information. So, I just entered a pennies trade that is sensitive enough to catch most tick moves and low enough to not overly effect my account should anything go wrong. Plus, I keep the trades frame open at all times.
I feel better already. I did also recall that I could read the balance patterns in a tick chart and also once found a tick series was key information toward a really big move within an hour after it. So, there's plenty of justification for this. I do need to close the penny trade if a real loss is possible, though.

mm ( post-ff 1y, other study, practice) compiled in Winner's Guide to Texas Hold'em Poker / Warren, Ken . Referring to it often, I held my account 24h, adjusting attitude to boot.

Updated pattern formula to something almost palatable: .

It works toward blending seamlessly spread trendlines and pattern areas.

I guess people'll avoid me like the plague, now, seeing that mess.
Ah, well.

Took a break from all the heavy mental lifting and tried a simple method:

Balance lines (drawing a ray (green) between two consecutive one-pip bars and, occasionally, when find a whole bunch of fairly close one-pip bars, a teal trendline) and Def-VSA_Spread indicator (whose colored spikes show trend and even bunches of one color show consolidation). Long as it's after a good drop, this works. Very simple, after all the lines and patterns I've done. And it's mine. Unusual.

Done well thus far:



1945EST Just waiting this balance between the high green and the high red (which chart above shows), looking for what'll break the tie. My guess is, eventually, a return to all those balance lines above has an above-50% chance.

1958EST Added a white line for a high-to-onepip spread trendline, and price kicked me into profit (in bright green. Then it turned where wanted, only was writing this, missing a far more solid move:



(not using heiken-ashi since not using ha patterns right now.)

2250EST Last trade of day ( http://i42.tinypic.com/t00mcm.jpg ) filled just below last peak, on channel above area returned to as reason) had large drawdown (~20-25%), gotta study it.

0452EST When I said 'study it', I was thinking of adding newsbar centerlines. I'll get to it in a moment. First, I wanted to show you the chart I awoke to, with useful lines highlighted in bright blue: http://i40.tinypic.com/ndvkhx.jpg . Impressive, huh?

Alright, now I have to figure how to get rid of old balance lines and retain the good ones. I already did highlighting, and opened a new chart without any lines except the grid.
I want this method to be easy, so I earmarked the last consecutive balance line I denoted last night, to start there again, from the old chart, and put it on the new, blank chart. Now I'll put consecutive balance lines on the new chart and see how it looks, before putting any of the highlighted or other spread trendlines on the new chart. Done: http://i43.tinypic.com/977csy.jpg .

Sticking to the new chart with its updated balance lines, I see the price is also.

I just lost a big chunk of change, margin call (33%) almost, ignoring mm.
Was going for the reversal off extremely rare one-pip bars. It did reverse, only locally and upon the new balance lines, not upon an imagined larger trend.

Reentered for same reason, stupidly counting on digestion since then. Guess must read book, again.
Chart added an 89sma centerline. http://i44.tinypic.com/2h4er8o.jpg

Alright, the chart's beginning to go my way, now. Perhaps balance lines direct the price on a new path, rather than suggest a turn for its rarity. One could see a two-pip bar for its rarity, somehow, in that case, when there's no one-pip bars to be seen. Instead, one'd look for another sign, such as the 89sma centerline, to trade around.

Now it's respecting the upper balance line on the last picture. That's bad for my trade.

The last one-pip bar was so important, in its own way, that it did a deviation: http://i40.tinypic.com/a0yr68.jpg . I guess I'd better get used to this obeying of the lines.

0554EST Didn't stand much of a chance. Looking at the old chart, PA was going with the white hi-one line. I'm exiting, without a place to exit. This is the difficulty I had last night on the drawdown in last trade.

1455EST I actually have the same account after six days. Six days!! Me!!! Oh, went thru several margin calls, but who's counting? I just did a sit-and-hold and I'm 24% to the good. There was a nice side with a horizontal ray (1.6293 g/u 07062009 0855gmt) from it, and I looked at it and thought, (07062009 2146gmt 1.6223) price's gonna return near here. And it did, well enough.

1929EST Lost account.
Why? It was, after ignoring mm last night and today, inevitable. Technically, I was evolving the balance line approach into a regular trendline method, so that I didn't trust the lines that were being obeyed right in front of me.
The old-line chart has been invaluable in displaying these lines.

@#$%%! Price returns. I wasn't there.

Again. It goes again. Toward my goal. Former goal. Aargh, overmargining....

I knew it. Pattern reading would've explained it. $%@#^$#. You'd have to understand what whatchamacallits are: See funny-looking piece of paper two posts ago, beside the hand-drawn head and shoulders; also, see post #6, again beside same. Head and shoulders patterns are already welll-known. Also, you have to look at the size and shape of each pattern. http://i26.tinypic.com/fwsilu.jpg http://i27.tinypic.com/nh13mb.jpg http://i27.tinypic.com/iwp43s.jpg http://i26.tinypic.com/jzv15c.jpg .
Note the patterns are made from regular candles, not heiken-ashi; ha candles are only more exact, since patterns tend to be more numerous in ha. I just happened to notice that the regular candles were providing more clarity at the time with their larger, sharper patterns.

method update:
Well, that's it. My trendlines are no longer the super-special lines, they're regular trendlines; they just have an extra kick of relevance. And, patterns are now essential to use; without them, I never would have figured the last trade., however wrongly I traded it.
Volume is already relevant, just kind of low-key (at least here.) Finally, well, I must reinclude news-bar centerlines, and in other ways relate in the above forms whatever reaction to the news is apparent.

-----

I hate to sic my pattern method on anyone. Hey, tho, if it works, why ignore it?

The area after the obvious stuff I showed above does the opposite: It's complex. I had to go into heiken-ashi mode, and discovered that bucket-like pattern that had repeated was flipped. You can see this well enough in regular candles.
What is more difficult to discern with regular candles here is that none of the patterns during or after the run are row three patterns, that is, whatchamacallits or head and shoulders. So I gotta explain what it actually is, as best I can, until I get it clear to others.
To review, I gotta explain all of it until others get the entire new pattern method, to which one adds my new way to draw traditionally-considered trendlines, as well as volume and finally an interpretation of news based on this. Plus, I have to keep doing mm to display its effectiveness in trading.

I looked at the regular-candle version of the current chart again, after reflecting on ha, and saw that it offered a little bit more information. The regular candle version is sharper; it just has extra, basic-pattern-irrelevant information, where ha makes plain where the pattern-relevant directions do or may be.

2115EST In the current chart, it seems there's a struggle between rl, an action of the contrast (circled c, on piece of paper) type, in row one, and a reaction, that of 'stacks', in row five. The rows have gone maximum distance from the center, or third, row; and, most of the microtrends are straight-edged, so any kind of a curve, or 'is' in rl-build, would have a strong reaction.
There are curves aligned with a portion of the current trend ("A"), which did trigger a recovery, and completion, of the 'stacks' idea. Rl then took over to resolve the top of the trend's peak with an odd spike at the beginning of it, which had declared a castle, rendering a challenge to stability.
Rl is next, and I'm trading on this educated guess, turning into a harami, to do the opposite of a peak and so resolve that odd-looking spike at the top of the trend. Or, it may turn into a whatchamacallit, returning row 3 to the view and, leaving the odd spike unresolved, leave rl's further discussion to 'stacks' resembling the extra shapes to the left and right of the central five.
It's doing the latter, and simply too, suggesting the trend's paying attention to the 'stacks' 's having several L patterns within it. This being the case, a counterweight around the now-central whatchamacallit pattern is to be resolved with a 2s balance pattern, and then a real turn will happen. I think.
It comes to be that it's actually doing a harami, only it's barely recognizable, being smashed against an unseen trendline and also busy copying small features of the trend above it.
Oh wait! I see it. It's gunning for a fighter jet pattern, which is a row-3 pattern that adds little features of a 'do', or straight-edged, pattern, to itself. So, it weighted on the row 4/5 side rather heavily, and now is doing an fj. Huh. Seems any hint of a row-2 pattern, which, to a rule, exactly counterweights rows 4 and 5 if used together, would flip the trend.

Fj's tend to flip within themselves rather fast, so I went ahead and exited my trade with the fulfilment of the shape of the harami ....
Hey, again. You always need row-3 imbalance, and a whatchamacallit followed the fj instead of some kind of row-2 shape. Now I'm worried, because PA broke a major trendline and I jumped on it. The other row-2 pattern besides the 2s is the vertical balance pattern. That means price may jump in my face, making the vertical balance pattern's midline upon the whatchamacallit's last piece.
2150EST No, it looked like it, only got a harami instead. I think a straight-edged version of a row-3 result/reaction pattern is happening. PA flattened on the major trendline, instead of breaking it, and that broke the 'equalness' of row 3 (for whatever reason, I got confused on the explanation for this), and took off like a rocket. A release, row 1's reaction pattern, and row 5 with its line segment action-'do' pattern, responded to the fj-whatchamacallit patterns dissolution into a row-three reaction that flattened. After this, a huge figure-s, which, I guess, combines two of the 'is' or perfect-type row-4 pattern, countered, with a little 2s pattern on either side of this 's'. A simpler introduction of a row 2 & 4 balance sometimes appears if row three patterns are dominant.
I would've lost the account on the second-guessing of the vertical balance, only I just margin-called, and made up the money in the next trade. So, only one account lost today. And that would have been made up, with only a little tweaking. Such is the forex business, I'm learning.
The pattern talk should get better; I was able to do so several times recently. It's not always easy, however, so that's why I avoided the method so much.

-----

0435EST 07072009 Boulder and collapse on the g/u 1' just-touching candles, right now. That's the row two area-type pattern. The chart was developing a loop in several different ways, and settled on this as it settled on an old trendline, before the trend dropped to a new, almost-parallel trendline.
I haven't made any new two - one-pip-bar trendlines since they began to get busy at 1600EST yesterday. I suppose it'd be worth my time to do it from g/u session open and on, what with no automatic indicator on it. Note in the picture below, which shows the bulbous curve copies ending in a boulder-and-collapse area, most of the trendlines are nearly horizontal, indicative of it being a while since the chart's had new trendlines put on it. Levels, not so much guides.



I'm just beginning on area interpretation as a pattern; here it makes sense, given yesterday's acknowledgement of there being on occasion very big patterns that fit into the general flow of patterns. So a boulder-and-collapse area, hmmm. Whoops, lost account. Hold on.
I took the margin call, which was a bit excessive. Okay. So I gotta bring in area-type patterns. That, with more trendline-to-pattern interpretation, I'd imagine. Then I can trade again. Alright. Makes sense.

-----

0840EST Lost several accounts. Gave up on mm, until figure patterns for real.

1110EST Okay, here's one. This brings the idea of area in more, exchanging with reaction and a little of action, not enough to engage the perfect-contrast struggle for balance nor that of curve, or 'is', vs. straight line, or 'do', both pairs of which increase complexity significantly when you have to combine them with the rest. Also, row pairings as a rule have 2 counter 4 & 5 or 4 counter 1 & 2, in a complex, high action chart portion. In more reaction pattern use, a simpler echo from row 3 is done, thus linking 2 & 4 and also 1 & 5. Thirdly, there's primary area pattern usage, in which case 1, 3 & 5 link and 2 & 4 modify their more-distinct patterns.

Been hemmoraghing accounts the past several days, and no go on the method thus far. Did come up with neat lines directly connected with news and deviations from it; it'll be something to show, once I get regular with it. Also figured patterns are essentially what I have on prt, only it's intimately attached to the cool lines; that's why they're so cool, they connect causation with all patterns' actions. Anyone can draw a line that looks sensible.
Problem is, I know patterns angle and pressure around these lines; I just haven't figured how to systematize it. Tonight, I, in my own personal thinking model, a representation of (any) self, I've worked on most of the day in spare time for 5 3/4 years, figured a shape to use within the thinking model that I can also use in the trading method.
I'm not sure it'll work yet. It's two parallel guidelines that boundary two line segments each with one end on one guideline and the other end on the other guideline, with the line segments never touching nor being parallel themselves. I stretch this device over a given pattern, aligning the two parallel guidelines around a crossed line, a group of patterns, or other relevant set of boundaries. The adjustable pair of lines explaining the rest of the pattern then guide the next direction in a logical series.



-----



A new VSA indicator is available. I did a fast innovation of how it could be used, having made no useful-enough sense of the author's handbook:





The above picture, which I'll clean up into set rules if it holds water and doesn't just blend into other portions of my method, establishes volume as explainably useful in my mind. In the picture below, price and volume get obviously linked together at places that define lines' pivotal changes:




This is one more step toward understanding why prt's patterns compress, angle and do other things, instead of just sitting there. I'm hoping volume will take care of this stuff, so I don't have to depend on some artificial setup such as the one I began this post with.

Sideways. First picture posted on ff about them, had so many one-pip bar trendlines I barely knew what to do with them. Yesterday, I could count on the lines I made from the last NFP. Still. After almost ten days. Then, late last night, the price reached a high compromise price, and consolidation has happened from then to now. A huge pattern is seemingly perfect, a consolidation extraordinaire, and it's taking in all energy from the usually highly-imbalanced symbol. Since late this morning, all one-pip bars have disappeared. Patterns copy wholesale, in sections and then individual ones. What lines are left are secondary news' bar's spread trendlines and a modified 89sma trend centerline. Get this, even the latter has gone for the copying individual patterns to make centerlines! Talk about breakdown. And forget about clean hits of lines, it's all compromise and skirt around price goals.
So my account's been all over the map while I learned this harsh truth about using two one-pip bars to draw my trendlines. That's okay, the overall "method", if one could be so premature about it, held my account together.

-1622EST Checked another broker's platform, and lots of one-pip bars showed there in the big perfectly-balanced pattern of most of today. I'm unsure what's going on, because there was no real correlation, say very few pips in the same bars of the platform I used as the one-pip bars in the other platform I just checked. It pays to think on your feet when you suspect corrupt data, trades that fail to go thru, a slow computer and the like. I'll include this notion from now on in the journal.

-0509EST, next morning One-pip wonders returned in style:

I've read RSO, almost replaced prt with a trendline version of the rows, and have been making large trades without much of a sweat, so it's probably time for me to post again. Of course, I can't do the new approach smoothly yet and I wouldn't doubt it'll change soon, and I still have large in-trade drawdowns and lose accounts regularly.

Breezily written, the new approach is organized, in one sense, on geometric points in space, that is, one point is a pivot, two a line, three a line and a point, four a frame or two lines, and five two lines and a point. In another sense, it is organized on prt's rows.

row 1a (two points, with row 2)
a 1-1 line that connects two related patterns at the initial area of the line and the current or a later particular area. This concept includes a midline with half of itself across a short, unbroken, straight trend (better seen in heiken-ashi.)

row 1b (three points)
A pattern, especially a roller-language pattern or one close to it, that crosses a spread trendline and reacts especially to it.


row 2 (two points, with row 1a)
a side, that is, several bars that follow each other and align highs and lows at about the same prices respectively. A 1-1 line deriving from a side would be included. Other variations, such as a horizontal 1-1 line may also be included.

row 3 (five points)
A complex organization of successive spread trendlines and the current price. The one I prefer is news lines, such as midbar centerlines, and lines that do major deviations from the news lines.

row 4 (one point)
Any area of the chart is filled with tiny angles, say 0, 45 and 90 degree ones, and have tiny straight or curved PA sections; one can see how these tiny spots play off each other to seem to move the PA line through a single pivot. Another approach is playing prt's two action patterns per row off each other, taking into account inter-row pressures and maybe even the occasional reaction pattern.

row 5 (four points)
A row's area pattern name can represent how major spread trendlines, more often the one from the last area and the one from the current area leading onward, are juxtaposed with each other in a given chart screen, or area; sometimes a major pattern is thrown in to connect the lines.


-----


1458EST072309 Also, I've gotten lazy. I keep depending on huge areas between very old lines. It works, but it's absurd. I could easily do better following the method above more closely. A good idea would be to act as if I'm posting on the Cable Update (GBPUSD) thread; there' a good bit of discipline to be had in terms of representing one's trades.

to-do list to adjust recent 1' candle/HA chart history to the last post's rows:

1) put in all 1-1 and news lines
2) identify central trend lines, usually only 2-3 (pair) per area, and put in bar channels that go toward this cause
3) figure copy, opp-copy, inside-copy patterns and their lines, and any original patterns and pattern sections that explain the central trendline path
4) pare work of unnecessary lines
5) describe micro-'s, pattern micro-'s (p/c's - is/do's) in latest complete and current partical areas

This list produces a chain of causality for any trade therein, and one should be able to read its meaning at any one point effortlessly. 1) is simple, 2) anchors, 3) includes, 4) simplifies, 5) redundant.
It should take about an hour per daily session, less after intraday breaks.

----

revision: I want to include reviewing the last trade for a containable number of psychological errors. I figured dealing with physical issues, such as gut instinct, emotional transfers from the chart's projected next move, and other, would do this.

1) Draw 1-1, news lines. (various, including row 2; row 3)
2) Define major trendlines per area. (row 5)
3) Include copied and typified-individual patterns with lines. (row 1a, row 1b)
4) Remove unnecessary lines and detail the last and current areas with (pattern) micro's. (row 4)
5) Correct physical of next trade, after chatting the largest period one can handle. (all rows)

It could take 2 1/2 hours to do the day's initial work, and updates of an extra 5-10' per trade.
The reward for all this work is being consistently profitable at high margin on the most pips available per type of trend / multi-area.

Time to study.

2011EST0726 Okay, so 1) Draw 1-1, news lines. I thought I'd go on and include 4) .
I seem to be getting perpendicular and short-trend related lines, with one small jump from a downchannel on Friday that continues today. So I'll cut down on the number of lines, thru choosing a wider space between consecutive one-pip bars to draw a trendline ray between.
This seemed to pay off, I got to a pattern copied once, both with 1-1 lines at Friday's 2002/2018 and 2036 / tonight's 2201 GMT , and drew a few deviation channels from midlines across these two lines. The channels hit the area near 0015 that rebounded into now.