Trading What I See

Setups are easier to trade when they occur at pre-determined locations. That’s why I mark off specific Fibonacci measurements as soon as a move extends beyond 162% of the first major pullback in a trend. Sometimes these levels will be ignored — but surprisingly often they correspond to pivots in the price action.

This morning provided a good example. In yesterday’s chart I had marked the next logical Fib level of 423%, and that’s where the market tried to stabilize. There was no assurance we would turn there, but look what happened between 6:45 and 7:00. The second bottom was a Fibonacci 127% external retracement of the bounce, and at the same time there was a divergence with the Stochastic oscillator.

That’s a nice buy setup, and since we had reached the 423% Fib level, my expectation was for a reasonable move. I first started watching for this level after reading Tony Plummer’s book Forecasting Financial Markets some years ago. He calls these levels Natural Reversal Points.

The concept involves market moves being influenced by the momentum of the next larger wave. Plummer considers a 262% external retracement of the first major pullback to be a “normal” move. As he says in his book:

…when a higher level trend counteracts the impulse wave, then the appropriate ratio for calculating the subsequent objective is 1.618; while if the higher level trend complements the impulse wave, then the appropriate ratio is 4.236…

And it worked again today. The Plummer book has the best explanations of Fibonacci relationships I have read. It is highly recommended, and is now out in an expanded version.

Where does that leave us now? The bounce continued for the rest of the day, but notice that the second major pulse reversed at 127% of the first pullback. And as price rose during the trading session, the volume contracted making the converging blue trendlines start looking very much like a WEDGE.

Nothing definite here, but it’s not really encouraging for any upside movement in the very short term.

Tony Plummer’s Forecasting Financial Markets

Yesterday I said that we had reached a logical level for a pullback. Actually I was anticipating an A-B-C pullback from the closing level followed by another five wave rally. Correct on the pullback — wrong on the rest. The “rest” was based on an Elliott Wave concept.

I no longer trade Elliott Waves, although many of the setups I use are based on them. I watch the patterns form and take signals from Fibs and trendlines if they match the patterns, but I don’t pretend to actually trade the waves themselves.

For those that haven’t studied Elliott, a reversal that starts as a five wave pattern (marked in yellow numbers between the magenta “X” and “A”) will usually have a pullback A-B-C (from magenta “A” to “B”) and then continue to make new highs for the move.

When Elliott Waves work as advertised, they are amazingly accurate. But when they don’t fulfill their promise I find that they have given me a strong directional mindset, and I lose more money when I’m wrong than if I just trade Fibs, trendlines, and chart patterns.

So instead of the rally I was expecting we ended up with an inside day. Since we didn’t exceed the 162% level I marked at yesterday’s close, my rally bias is no longer in effect. But you’ll see from the chart that some good setups still appeared.

Much of my knowledge of Fibonacci came from studying Elliott, and I recommend that traders at least have a general idea of the structure of impulse waves. I’ve included a link below to the most understandable of the many books on Elliott Wave.

Robert Prechter’s Elliott Wave Principle

Both yesterday and today we set up five point ranges that could have been called rectangles, but if you’ll check your technical analysis books you’ll see that the volume is supposed to significantly decline during the pattern. That isn’t happening.

This morning we dropped to the bottom of yesterdays “pattern”, came back up to the top and “broke out.” And reversed. That was a potential trade, turning down at the 127% external retracement of the morning drop. It matched a divergence in the Stochastic, and formed another Wyckoff Spring entry.

You’ll find many examples of the Spring using the search box in the right panel, and the most recent one happened yesterday. They are basically breakout failures. When the market breaks out, it is supposed to continue — not reverse back through support or resistance. It often leads to a tradable move in the opposite direction.

The exit was yesterday’s low where we started sideways again. Notice the divergence that said further downside activity was unlikely. Divergences don’t always work, but when they confirm some other potential turning level it’s a good idea to pay attention.

For the last week I’ve been pointing out how important the Fibonacci A-B-C pattern is for finding turning points in the market. It almost looks as though the Russell 2000 wanted to emphasize the issue with today’s trading.

Yesterday I showed how the larger A-B-C pattern had just reached the 100% measurement (with a reminder that it “does NOT necessarily mean we will bounce on tomorrow’s open.“) But that was a logical level to be watching for a bounce.

We opened with a small pop to the upside and reversed to break yesterday’s low. Are we going lower? We’re still at the large 100% A-B-C shown yesterday, and if you check the measurements using the drop to the close as a smaller A-B-C, the market reversed at 162%. That low was also a 127% external retracement of the “A” to “B” move while setting up a nice divergence with the Stochastic.

That would be enough to take a long position at today’s bottom, but to give some added assurance, the entire formation set up a Wyckoff Spring reversal that I mention from time to time. That always translates into a close initial stop.

Measure another A-B-C move to today’s top, and we turn down at 162%. Add some more A-B-C patterns and the market has created a nice rectangle that contained prices for the rest of the day.

Many of you are probably wondering whether the A-B-C pattern is just an Elliott Wave correction. There is actually a theory that Elliott had it wrong, and that most market movement (even Elliott’s five waves) can really be explained by a better understanding of the Up-Down-Up pattern that I call an A-B-C.

I won’t take sides in that argument, but I often get Elliott counts wrong while A-B-Cs never change their structure. I developed much of this method from Tony Plummer’s book Forecasting Financial Markets. He recently came out with a new edition. Check out the link shown below.

Tony Plummer’s Forecasting Financial Markets

For some reason my data provider (eSignal) always includes several hours of holiday trading in their feed, and although I can screen it out in the S&P Futures, it still shows up in the Russell 2000. If you can ignore the “extra” bars on the chart, this morning’s bounce came at a logical Fibonacci level.

Without the extra data the trendline (from the 15 minute chart) would hit at a slightly different level, and perhaps there would have been a divergence in the Stochastic — there was in the shorter Stochastic I was watching at the time.

We opened with a small gap and a 15 minute drop (A), made a three bar pullback (B), and bottomed (C) with a slight penetration of a trendline that started from the February 12 low. The bottom was a precise 78% retracement of the rise from the low on Friday.

Whenever I have and A-B-C move (down-up-down) I take the distance from X to A and multiply it by Fibonacci ratios to get 62%, 100%, and 162%. These numbers are subtracted from the pullback (B) to give potential turning points for the move.

If the move is strong it will carry to 162% or further. If it is an average move, it will stop at 100%. And if it turns at 62%, you are more likely to get a good reversal. Trendline support, a 78% retracement, and an A-B-C that turned at 62%. We got a much better reversal than I expected. There was no reason to exit until we got to the rectangle area at lunch hour.

I don’t trade rectangle breakouts unless there is a nice volume increase, but the pattern on this one looked good. A breakout at about 11:00 (Pacific) followed by a pullback to the support at the top of the pattern. The first target on a rectangle is when price duplicates the width of the rectangle. That was the high of the day.

Some days the market will only give me half the signal I’m looking for, and that’s what happened today. Fortunately half a signal isn’t enough to trigger a trade, but it certainly can be frustrating.

After a one-bar pop above yesterday’s high we dropped for about 30 minutes and then put in the low of the day. The bottom was only a 38% retracement and lacked a divergence, so I didn’t have an entry.

As soon as we made a new high I followed my usually procedure — marked off the 127% external retracement and the potential targets for an A-B-C move (shown in yellow.) We went right through 127%, and then the mid-day top occurred between my targets. The fact that we had a nice divergence at the Test of Top wasn’t enough for me to trade without the encouragement of some Fib levels at the same time.

The afternoon pullback made another A-B-C pattern with the length of C equal to the length of A (a Measured Move.) Notice that it also retraced right to the support of today’s opening. Two reasons to trade, but I wanted a divergence to make it three.

With two reasons for a bottom at 11:30, why didn’t I take that trade? Because although I don’t have any statistics to back me up, I feel that a lack of early setups is too often followed by failing setups late in the day.

Yesterday the signals were clear and worked. Today we were out of synchronization. In fact, even the various markets are providing conflicting information. The Russell is the only one that closed at an intraday high. Is it leading the way, or just out of step? That is going to make me cautious at tomorrow’s opening.

I wish the Fed would make its announcements over the weekend instead of messing up a perfectly good trading week. But even on Fed day the Fibonacci levels keep working. I just question whether it is worth trying to trade them.

We started with a fast drop that reversed right at a 127% external retracement of yesterday’s last rally. It was no surprise that we then went into a sideways pattern as everyone waited for the Federal Reserve Policy statement.

I’ve marked two A-B-C patterns — one in yellow and one in blue. This type of overlapping pattern will often end in a Fibonacci cluster with several measurements pointing to the same level, in this case to point “C.” On another day this might be tradable, but I took a long walk about that time.

There is one interesting feature today that is sometimes quite useful. Although I first heard about it from Robert Prechter (Elliott Wave Principle), it is best explained in Connie Brown’s Technical Analysis for the Trading Professional, one of my favorite comprehensive trading books.

Notice where the two trendlines that create the triangle pattern cross. After a triangle breakout the market will often change direction right at the triangle apex. The top came very close to a Fibonacci 162% of a even larger A-B-C pattern, so you would have had both a price and time target. Always useful information.

Connie Brown: Technical Analysis for the Trading Professional