Trading What I See

Once again we got through the first hour without a setup that fit the trading style I show on this blog. Unfortunately, during summer trading that happens much more often than I would like.

It’s not that there are no possible trades on today’s chart. It’s just that when trading in congestion I’ve found that I have many more losing trades that tend to wipe out all the gains from the winners. So I’ve developed several methods to help me define congestion, and learned to live with missing the occasional strong move that sometimes occurs on a breakout.

I define congestion as times when the ranges of a long series of bars overlap that of a single bar. Often I’ll use a longer time frame to watch for this situation. Here’s a post from last year called Overlapping bars equal congestion that shows what I mean.

Back in February in a commentary called Non-Trading Tips I showed a two-day period where, on a fifteen minute chart, almost every bar in the trading day fell inside the range of the first bar. That’s extreme sideways action, and my avoiding trading yesterday and today are just extensions of that method.

Where there is no tradable action in the first hour, unless there is a runaway move, I assume we may be starting a period of congestion. When I mark off that first hour range, I’m just simulating a longer chart on top of my trading chart.

If you’ve followed the earlier links and understand the general concept, take a look at a 45 minute (or hourly) chart from today and you’ll see that EVERY bar today touched the range of the first bar.

You need to decide whether breakeven trading during congestion is worth catching the occasional good move. Personally, I’d rather wait for the stronger setups during intraday trends.

Evidently the market decided it wanted to spend some more time working inside the daily rectangle. In fact, today we dropped back almost to the center of the pattern. With a gap down opening, price built a floor just under the 850 level and went sideways until after the lunch hour.

When there was no trade in the first hour, I marked off the range and waited for a breakout (A) and a possible pullback (B.) Then I passed the trade that turned into a Measured Move.

On pullback trades I want to see a decrease in volume before I’m comfortable with a trade. Sometimes volume doesn’t make a difference, but over time I get better consistency when I insist on either correct volume or multiple reasons to take a trade. At least I got a lot of paperwork done today.

After opening with a small gap above yesterday’s high, the market immediately reversed on the first one minute bar and closed the gap (not shown.) That looked like a Spring setup, but there was no confirmation — no Fibonacci level or divergence for the top.

The next two minutes reversed the reversal, and then the market topped out at the 127% Fibonacci external retracement. For those new to the site, that’s a measurement that retraces more than 100% of the previous move, shown on this chart with white arrows early in the morning and again in the afternoon.

Any time the market exceeds a previous pivot you should mark off the 127% and 162% external retracements, since these are the two most likely levels for a quick reversal. If we exceed 162% that’s a good indication that the move will continue for a while.

The turn at the high created another Spring setup as we broke through yesterday’s resistance but immediately reversed. This time the reversal occurred at 127% so there were two reasons to enter a short sale. If you missed that entry, the pullback at “B” was at a 50% retracement of the early decline. I like to enter after Dunnigan blue reversal bars if there is another good reason, since that often acts much like the Spring setup.

An A-B-C pattern (shown in blue) will often turn at 62%, 100%, or 162%. The two bar hesitation at 7:30 was at 62%, but there was no actual bounce. We went through 100% without a pause, and finally made the low of the day at 162%. Nice morning trading.

For the next four hours I can’t find an entry I like. I expect early pullbacks to be 50% or greater, but it never happened. I have a nice parallel trend channel marked in yellow, but without an entry you can’t trade it.

But it does point out the afternoon top, which is also a 162% external retracement and another refusal to move upwards after again breaking yesterday’s high. Not as clear as the morning reversal, but I would still consider it a Spring.

You might take a look at the daily chart before tomorrow’s open. We’re back at the high of a rectangle that has lasted over a month. For longer term traders tomorrow might be decision time.

The market made an immediate reversal just after today’s opening, and declined for the next 30 minutes to the eventual low. Without any pauses there wasn’t much chance for a safe entry. And without any pullbacks there were no internal measurements to pinpoint the bottom.

So even though we turned at a 78% Fibonacci retracement level, by the end of the first hour I had found no tradable setups. That’s when I drew the magenta bars to mark off the first hour’s high and low, and although we broke out of that range there were still no clear entries.

Finally at the pivot marked with a circled “4″ there was a potential entry. As I said yesterday

where there are two pullbacks in a move, the first one will usually be large (most often 62%) and the second one small (38%.)

Of course the movement over the next hour was just sideways so I gave up on the trade. I should have remembered that we are now into the summer trading season where many of the moves seem to happen in slow motion.

During this season it’s often useful to pay more attention to (but not necessarily trade from) the longer term charts because many of the patterns are extended. Even a move from a three to a five minute chart can clarify moves, since you can include more of the previous day’s trading.

By including part of yesterday’s move on this five minute chart you can see the structure of an Elliott impulse wave. If you realize that Wave Four is often a long drawn-out affair, it might increase your patience in this type of a trade. Remember that I don’t actually trade Elliott Waves as such, but much of what I present on this site is derived from Elliott principles.

There’s another reason to illustrate an Elliott Wave today. Last night I received an e-mail from Elliott Wave International. When they found some of my Elliott comments on Trading What I See during a Google search, they asked me to become one of their associates.

Doing so allows me to put up some links (left sidebar) to free information they offer on their site, such as a very nice 10 lesson Elliott Wave tutorial. For those of you interested in Fibonacci and market structure I highly recommend it.

If you follow any of those links, you’ll have to sign up for Club EWI which puts you on their mailing list. Fortunately they don’t send out material excessively. But it’s a no-cost way to learn more about how the market works (at least some of the time.)

Yesterday’s market showed multiple parallel lines, but even a single parallel will often confirm a target. This is particularly true when price breaks out of a triangle.

Last week ended inside a small triangle (magenta lines), and today’s first bar made the third touch of the pattern bottom. Triangles without decreasing volume have a much greater failure rate, but as you can see with today’s action sometimes they work anyway.

There are several ways to create a target for a triangle, but the simplest is to draw a parallel to the side of the triangle opposite the breakout. Price will often stall or reverse when it reaches that line as it did today. But targets are not necessarily reversal points — just places to take partial profits.

Today is a good example of how I continually change my trendlines and parallel channels as the day progresses. After the breakout there was a small pullback that allowed drawing the lower yellow trendline. The pause at the triangle target ended as we hit the extension of that line. As soon as we continued higher, the yellow parallel could be drawn, and the next pivot top occurred at that parallel.

Once again we went sideways instead of reversing, and if you looked at a one minute chart you would see a small down channel could be drawn using parallel lines. When we broke through the top of that channel the upmove continued, and a new trendline (blue) could be drawn through the pivot at “B.”

A parallel to this blue trendline also confirms a logical turning point just as price reaches a Fibonacci extension (C=62%.) I draw most of these lines during the day, but often erase many of them to reduce confusion when I post a chart. Even the parallels that are never reached give important information about the momentum of a move.

A book I often recommend (Edwards and Magee’s Technical Analysis of Stock Trends) gives step by step directions on how to draw these ever changing channels that help guide you through the trading day. Add a little Fibonacci and you get confirmation of signals from two entirely different methods.

Edwards and Magee: Technical Analysis of Stock Trends

You’ve probably noticed that recently I seem to be showing a slightly larger picture of market action. Actually I try to look at this every day, but since right now I can’t follow the market in real time (and may not be able to for another month), the larger picture is what I see first. And it has been providing interesting information.

On Thursday I pointed out the Trader Vic 2B Top. They don’t always work out, but since they allow a close stop, I never ignore them. They are, of course, a specialized instance of the Spring patterns I like to trade. Since the pattern took over a day to form, it was a warning that the larger trend had probably changed to down.

Friday and today completed what is called a Downside Triangle with a downside break after about four hours of trading. Triangles of any type usually break with a strong price surge, and today was no exception. One thing I watch for in all patterns, but particularly in triangles, are internal Fibonacci measurements. The better the Fib setup, the more likely the triangle is to work as expected.

I have three peaks numbered in the chart representing the three rallys making up the triangle. The second rally is 62% of the first, and the third rally is 62% of the second. That type of symetrical pattern seems to make the breakout more likely to reach its minimum target. In this case the target is the parallel to the upper edge of the triangle which we hit near the close.

The triangle itself is made up of an A-B-C pattern which has passed the two first logical Fibonacci targets. Notice how it tried to bounce at 100% and 162%. If it continues, the next target will be 262% which usually creates a bottom. And if it develops into any kind of panic move it could drop to 426%. Check out a longer term chart and you’ll find old congestion near these levels.

Which doesn’t mean we won’t turn around tomorrow morning. Fibonacci targets are not predictions, but if you’ve read this commentary for very long you are aware of how often they work out.

Trendlines form channels and chart patterns, but in addition they often give indications of the strength or weakness of a move. Converging trendlines usually indicate weakness, or even a potential reversal.

Yesterday we closed at a down sloping trendline, and this morning reversed direction with a small gap. Today’s chart shows all of yesterday’s price action, and you can see how the yellow trendlines converge. That was an early warning that the selloff was losing strength.

There were three reversal indications at the bottom. First we hit the lower trendline. Second, price had reached the 127% external retracement of the last rally, which is often a turning point. And third, the reversal formed a Spring by breaking a support area and immediately reversing. A divergence would have been nice, but not really necessary.

The yellow trendline again formed temporary resistance, but then we broke through with a slight increase in volume. If you missed the bottom reversal, the pullback to the trendline (and Fib 38% retracement) was another potential entry. Then the market started giving conflicting signals.

When a first pullback is only 38% that often indicates strength. But when the next turn is only at a 62% extension (C=62%) that shows potential weakness and will often mark the end of a move. But then the blue uptrend line held, and the A-B-C became a Measured Move with the final B-C move equaling the original “A” rally.

By this time the blue converging trendlines were obvious, and the pattern looks like a WEDGE. A breakdown from a wedge usually retraces the entire pattern, but instead it just moved sideways for the rest of the session.

End result — we regained yesterday’s loss on slightly increased volume. But we did it in a manner that that wasn’t very convincing. I guess tomorrow I’ll just have to take it one trade at a time.