Reading Charts in Real Time

[thalestrader 14]

Every now and then someone runs the figures in an open forum, I can't recall any that have demonstrated an edge with 123's taken alone. I wonder what you are basing your it "provides a good statistical edge" argument on I would be very interested in the other money management parameters you used to reach this conclusion?

 

I've never been very proficient at back testing anything other than systems that somehow reduce price to a product/sum/quotient/difference and assign a value based upon the calculation of said equation, e.g. a simple MA crossover system, an OB/OS system using an oscillator. I have no idea how you back test for this [other than to define a 123 by saying if price makes an nBar high, followed by an (n-3) bar low, followed by a (n+m) bar high </= nBarHigh and makes a Low<(n-3) then sell]. In other words, how are you defining a 123 for back testing purposes? I make no claims to be a systems/testing expert, and I really am simply asking the question.

 

Otherwise, I do not see how this approach cannot provide an edge. Certainly, context will sharpen the edge. But even on its own, so long as it is uniformly applied as respect to the degree of the swings, how can it not provide an edge? Every trend change starts with one. And the risk/trade is narrowly defined, so even when one does get the ABC correction that sucks him or her into what turns out to be a mere counter trend correction rather than a reversal, the MM should keep one form any catastrophic loss.

 

Best Wishes,

 

Thales

[Cory2679 11]

Anyone who uses or has used NinjaTrader, please check out the thread I just created...

 

http://www.traderslaboratory.com/forums/f208/ninjatrader-gain-missing-data-7403.html#post86459

 

Thanks!

 

EDIT: This is NOT AN ADVERTISEMENT!! I NEED HELP WITH SOMETHING!!

 

(After re-reading my post, I thought it may have sounded like an ad. )

Edited January 14, 2010 by Cory2679
Clarification

[Marko23 10]

...

 

Re: backtesting.

 

I have to say, I prefer trading.

[thalestrader 14]

Re: backtesting.

 

I have to say, I prefer trading.

 

You and me both, brother!

 

In the end, I found back testing to be quite useless for me. But I chalk that up to me not knowing what I'm doing.

 

Best Wishes,

 

Thales

[Cory2679 11]

Current look at the EJ.

 

I'd be expecting a long opportunity any time now!

 

(There is a similiar situation with the EU and maybe even the UJ as well.)

 

-Cory

[thalestrader 14]

6B sell stop 1.6277 targeting 1.6202, but I'll be watchful at about 1.6247 or so.

 

Best Wishes,

 

Thales

[thalestrader 14]

6B sell stop 1.6277 targeting 1.6202, but I'll be watchful at about 1.6247 or so.

 

That's too much back and forth now, and if it going lower, it will have either to go higher first, or go lower without me.

 

Best Wishes,

 

Thales

[thalestrader 14]

Current look at the EJ.

 

I'd be expecting a long opportunity any time now!

 

(There is a similiar situation with the EU and maybe even the UJ as well.)

 

-Cory

 

Depending upon your market maker, price either tagged or stopped just above the 423 expansion of the first leg down off the high, i.e. 423% of 1-2 as measured from the high of 3 using JR's terminology.

 

 

So, this would be as good a place as any for the bears to take a nap and the bulls to try and push back. 131.25-131.50 looks to be a level at which something important could take place.

 

 

 

Best Wishes,

 

Thales

[Marko23 10]

Eur/USD found support at 1.4445.

 

ADD: 1H chart added for support. I'll quit.

Edited January 14, 2010 by Marko23

[Cory2679 11]

I'm done for the day. I'll be honest...I didn't have a single trade this morning!!

 

If I would have been up early, there were some obvious great short opportunities today.

 

It's all right, though. I'm doing pretty well this week...as long as I don't screw it up tomorrow, I'm really happy this week...I'll post my week's P/L on the P/L thread tomorrow after I finish trading.

 

But anyways, on a random note: not that anyone necessarily cares but I thought I'd say this anyway: I don't post every trade I take onto this forum. I only post when it is convenient, and I make a point of not making a post if someone else has already made a post concerning that chart (that is relatively similar to what I would have posted).

 

Here's a last look at the EJ...not taking this but I might if I were going to be here and it wasn't so late...

 

NOTE: Forgive the chart, there's actually a break in the data between the low and the higher-low (it's not really that noticeable on this chart, though)...I've been having problems with Ninja today...

 

-Cory

[Cory2679 11]

Still not trading...leaving any minute...but just for fun...here's a look at the GU...

[Cory2679 11]

Last post and then I'm out...

 

This is the weekly chart of Bank of America.

 

My girlfriend's parents bought this (cheap---around $5) and I believe pretty much have the "buy and hold forever" mentality with it.

 

Obviously there are infinite arguments for either side of whether the economy's really taking off or if it's just a bear market rally.

 

SO, I thought I'd look at the chart...

 

If I were "trading" this, I believe I would just be waiting....waiting for price to break that range. I've outlined the range by my horizontal lines, and pointed out why the levels are significant with rectangles and ellipses.

 

Any thoughts??

Edited January 14, 2010 by Cory2679

[Marko23 10]

...

As an aside I also look for these reversal patterns at HOD's and LOD's as that seems to be quite good context as well. Does anyone else find the same thing?

...

 

Yes, in my future trading I follow 3-day highs and lows and for EU I observe a similar rhythm. Therefore a 123 down after a new 3-day high is a chance for short for me. Yesterday I was a bit cautious.

[thalestrader 14]

Here's a last look at the EJ...not taking this but I might if I were going to be here and it wasn't so late...

 

 

You'd be off and running and close to PT1 if you had traded this...

 

Best Wishes,

 

Thales

[thalestrader 14]

I'm kicking myself for not having snapped a pic of this ending diagonal on the GBPJPY before it broke loose, but if you have bar replat mode, you should be able to see it well before the fact.

 

Best Wishes,

 

Thales

[TradeRunner 10]

Monday: -75 (A bad day)

Tuesday: +45 (Lucky to end positive)

Wednesday: +30 (Should have been more)

[MidKnight 12]

Otherwise, I do not see how this approach cannot provide an edge. Certainly, context will sharpen the edge. But even on its own, so long as it is uniformly applied as respect to the degree of the swings, how can it not provide an edge?

 

I think BFish brought up great points and points that are no stranger to this thread either. I know you are unable to see how it cannot be an edge, but you are looking at through the eyes of experience, so of course you are going to say that. I think there is something to it still, but it is a discretionary feel about it, rather than something I would say on its own is an edge.

 

 

Every trend change starts with one.

 

I need to run out to do some errands, but later on, I will post several 'trend' changes that do not start with a 123 reversal, when measuring trend as per PA and comparing across the same degree.

[Kiwi 16]

Re: backtesting.

 

I have to say, I prefer trading.

 

This edge for the 123 (break of pullback to hl or lh) discussion is reminiscent of one at FF on inside bars. And the funny thing is that its backtesting that allows you to see that there is little or in some cases no edge.

 

Here's the thing. The backtesting is actually right. If you take every 123 with a 1.1 to 1 target to stop then it will have no or little edge. Same with Inside Bars. Same with pinbars. Etc etc. One of the things in John Hill's book is that he does discuss the edge of various patterns/bars.

 

 

I wrote a post about various types of entry and purposes here and don't want to repeat it.

 

 

The unique edge for a 123 is that it gives you a well defined entry (price continuation after congestion following the probable start of a move) and a "tight" stop on that congestion rather than the bottom below it (if long). But by itself its pretty useless - its only useful in defining a good point to take a risk in the bigger context. Its edge over a price bar breakout in a pullback is that there is "more" continuation indicated by the break - but its at the expense of a bigger stop so different traders have different preferences.

 

Similarly IBs in the DIBs method at FF. They test horribly. But they form at the early point in a bigger timeframe trend and give you a high probability in that context.

 

The problem that MK showed is that when he took all the 123s he liked, his context wasn't good. And when Thales took all the 123s he liked his context was good. And someone else might be in between.

 

So I would counsel those following the thread to look at two things:

- how does thales use the 123 for entry and as a basis for managing his ongoing trade?

- how does thales pick up the context that really makes it sing for him?

It is real and it is there - but you'll have to work at it (read those two books!) and it will also have to be something that fits what YOU can do. Why do you think there are so many different approaches to trading? Why do you think so many fail trying to do what others can do? Even when they see it laid out in front of them in detail?

[BlowFish 10]

I am glad you asked that question, because it is a great way to let me make some general points that I think may help clear some things up for folks regarding this approach.

---snip---

 

Thank you for taking the time for such a detailed and well thought out reply. One of your most lucid to date (imho)! Getting late here (a wee bit past midnight) so I will read it again more carefully in the sober light of day (with a 240min chart alongside too). You really have a good eye for PA and pulling apart some of the trades I hope will reveal some of the nuances. In some ways it is academic....a long would not have triggered and if one had passed on the short....well another willl be along before long. This one had me chuckling though, good trade.

 

Incidentally whenever I use a word stronger than 'anticipate' (like think) ....'anticipate it might' is what I mean. I don't slip up too often but now and then:)

[BlowFish 10]

 

___snip___

 

Otherwise, I do not see how this approach cannot provide an edge. Certainly, context will sharpen the edge. But even on its own, so long as it is uniformly applied as respect to the degree of the swings, how can it not provide an edge? Every trend change starts with one. And the risk/trade is narrowly defined, so even when one does get the ABC correction that sucks him or her into what turns out to be a mere counter trend correction rather than a reversal, the MM should keep one form any catastrophic loss.

 

Best Wishes,

 

Thales

 

This is probably not the thread to get into a drawn out discussion on this. Im happy to leave the clever stuff to Kiwi I will make a couple of 'PA' type comments, If you 'hand test' by marking all the 123's one of the issues sis that you will often get '3 legged' (ABC) corrections in a trend ,these will result in a 123 that will likely fail ( a JR one will though a Vic one migh not 'triigger'). You also often get them in congestions and they tend not to work out so well. Again I should be clear I was talking about 'JR' 123's. Trader Vics have a trend line break as an intrinsic part of the pattern if my memory serves me correctly? This makes them much harder to test automatically but arguably makes them more 'robust'.

 

Mind you JR cautions traders about looking for them everywhere. Actually most of his material is aimed toward trading the first pullback after a 123 (the infamous hook). So the 123 is simply the precursor to the hook. Really should go to bed now.

[patrader 10]

Edited January 15, 2010 by patrader

[thalestrader 14]

Thales's Secret? Not secret, but probably neglected.

 

Blowfish, you have been very helpful to me today, as you have provided me again with a perfect opportunity to make a point that I want to make, this time in response to Traderunner's charts of the last three day's of his trades. (Please don't think I'm picking on you, Traderunner as I do truly admire your efforts). In fact, it is a point that is too often overlooked here, and perhaps so obvious to me I fail to mention it as often as I should.

 

This is probably not the thread to get into a drawn out discussion on this. I'm happy to leave the clever stuff to Kiwi

 

I think both you and kiwi are both very clever, and I am feeling the heat of being on the other side of the trade from both of you on this one, so I am doing some thinking and rethinking - in addition to being terribly unoriginal, I do think I have a fair degree of open mindedness. I am very happy that you both have been willing to share your insights here - I for one have learned new things from each of you.

 

 

Trader Vics have a trend line break as an intrinsic part of the pattern if my memory serves me correctly? This makes them much harder to test automatically but arguably makes them more 'robust'.

 

Yes, Vic insists on trend line breaks. In fact, the "1" is not strictly speaking the high, but the break of the trendline itself if I recall. But you do not sell the break, you wait for the retest of the high, then sell a lower low.

 

 

Mind you JR cautions traders about looking for them everywhere. Actually most of his material is aimed toward trading the first pullback after a 123 (the infamous hook). So the 123 is simply the precursor to the hook. Really should go to bed now.

 

And this is precisely the point I wanted to make in response to Traderunner's trades. On any given day, there may be 1-3 good solid opportunities on the 6E, the 6B, and usually only 1-2 on the 6J. From what I have seen of the EURJPY, I would say 2-3 good opportunities there as well during a 24 hour period. So when I see someone posting charts with 5-7 trades in one day, I would say that there is some over trading taking place.

 

I think your point concerning Joe Ross's caution is extremely important. Back in November I posted the following:

 

... I am also influenced by Gann. Gann once said that the most profitable place to enter a trade is after the first reaction off of an important high or low. Not an earth shaking concept, by any means. But if you think about what I am doing with respect to the H-L-LH and L-H-HL, I am looking to enter a position after the first reaction off of what may be an important support (bottom) or resistance (top).

 

If you look at the trades I post here, I am always (at least, almost always) looking short after a rally or long after a decline. I will add to a short in the middle of a decline if I am already short and price indicates it has not spent its energy. I will add to longs in the middle of a rally if already long. But you do not see me looking to buy the second 123 or sell the third 123 as we sometimes see posted here. In my mind, I got one shot to get into a move. If I miss it, I miss it. I'll catch the return trip.

 

I would only add that for every rule or guideline I post here, I have a bushel basket of trades in which I broke with one or more. But I rarely break with this one.

 

Best Wishes,

 

Thales

[thalestrader 14]

...

 

Excellent!

 

Best Wishes,

 

Thales

[Gabe2004 10]

when you get a large move in either direction with obvious momentum behind it, do not fade that move.

 

 

If you look at the trades I post here, I am always (at least, almost always) looking short after a rally or long after a decline.

 

If OBVIUOS MOMENTUM = RALLY/DECLINE then how do we know in which case they are not equal?

 

Thanks

 

Gabe

[thalestrader 14]

If OBVIOUS MOMENTUM = RALLY/DECLINE then how do we know in which case they are not equal?

 

You've mentioned this twice now, Gabe, but each time you snipped this phrase but leave out the context.

 

I would not have taken the short where you did, as price had rallied in three waves with most of it during that extended third leg. I know Elliot Wave gets a lot of bad press, and I try to keep it to a minimum here as I know there are many who would discount this whole approach based on the introduction of Elliot. But, when you get a large move in either direction with obvious momentum behind it, do not fade that move. Wait for a choppy pullback and then a push to a new high that stalls and presents a clear 123 going the other way.

 

Perhaps I should have said "if you get a choppy pullback, wait for a push to a new high that stalls and presents a clear 123 going the other way."

 

Here is the chart price action in question. The area in the first rectangle appeared very choppy, sloppy, and overlapping. When I see that after a move with obvious momentum begind it, then unless I have a real big picture S/R level right there, I am inclined to wait for another push higher (or lower, in the case of a short) and wait for a more clear 123, which I think the price action in the second rectangle represents.

 

 

 

The reason for waiting is that if I can count three swings, and price does not seem to commence a sharp swing reversal, then price is likely to push for one more swing, making the entire move a five wave affair. I do not make much of Elliot Wave theory here, but as I have said, I make use of it myself. This is one of those places where I find it helpful.

 

 

Best Wishes,

 

Thales