The Three Hard Rules of Elliott Wave Theory
October 2nd, 2008 by Corey Rosenbloom
Despite how difficult Elliott Wave Theory seems - and it’s certainly not easy to interpret possible wave counts - there are really only three solid, objective rules that your wave count must follow - the rest is mere guidelines. What are these three main rules you ask?
Let me preface by saying that Elliott proclaimed that Impulse or Motive waves unfold in five wave sequences (three up separated by two corrective waves down) which forms the core of Elliott Wave. Interpreting corrective waves can be endlessly difficult, but it’s wise to start with the initial Impulse Structure and follow these unbreakable rules:
1. Wave #2 Cannot Retrace 100% of Wave #1
2. Wave #3 Cannot be the Shortest Wave (but does not have to be the longest)
3. Wave #4 Cannot Enter the Territory of Wave #1
Let’s look at these closely:
Rule #1: Corrective Wave 2 CANNOT Fully Retrace Wave #1
Generally, we look to Fibonacci retracements in terms of impulse moves - especially to 31.8%, 50.0%, or 61.8%. If you try a Wave Count and your possible Wave #1 fully (100%) retraces what you think is Wave #1, you need to re-lable Wave #1 and start your count over. Often, it’s very difficult to trade real time in developing Wave 1 structures.
How does this look in an idealized example?
Wave #1 is Green and Wave #2 - the Corrective Wave - is Red. The Red Dotted line signals a wave ‘danger zone’ and if Wave #2 crosses the blue dotted line (the start of Wave #1), then the Wave Count violated Rule #1.
Often, Wave #2 will be a deep retracement of the first wave, and can retrace to 61.8% of the prior wave or slightly beyond.
Rule #2: Wave #3 CANNOT be the Shortest Wave
Often, Wave #3 moves are traders’ dreams, as they contain sustained price action in a given direction. These are also the zones of the “Sweet Spot” structures in price moves as well.
When I was first learning Elliott, I though that Wave 3 had to be the strongest, longest wave - not so. It just can’t be the weakest/shortest of the impulse waves. If wave #3 seems week, it could signal a stronger Wave #5 is possibly yet to come.
What does this rule look like?
I’ve highlighted Wave #3 in this example. Wave #1 is the shortest in this diagram, and Wave #3 is slightly longer than Wave #5 - this fits the criterion.
If you’re trying out a Wave Count and what you label as Wave 3 turns out to be shorter than your Wave #1 and Wave #5, you need to start again and re-label your chart according to Elliott.
Finally, Rule #3: Wave #4 CANNOT Enter the Price Territory of Wave #1
Wave #4 is known as a profit-taking wave, and as such, shouldn’t give back a large portion of Wave #3. Something’s wrong if Wave #4 is a very deep retracement. Let’s look at this on a diagram:
Wave #4 can be expected to retrace perhaps 31.8% to 50.0% of Wave 3, but start to worry if the retracement gets deeper than these levels. I’ve compressed this idealized image, but if what you think is actually a Wave #4 correction, you could place stops just beneath the top of Wave #1 - the black dotted highlighted line in my diagram. I’ve drawn an “X” at this level.
These are just the surface level basics of Elliott Wave Theory. We’re having to go deep within the concept, perform counts on multiple charts and multiple markets, and read multiple sources for the Chartered Market Technician (CMT) program from the Market Technician’s Association ( www.mta.org). It’s eye-opening and perspective broadening and very interesting. The information above comes from these sources, which also can be found on a variety of sources online including the Elliott Wave International site.
I’ll continue to try to share insights and basic lessons learned.
Don’t try to become an expert overnight if you’re interested in Elliott Wave. However, if you do want to try it out on a few charts (it tends to ‘work better’ for broader indexes and markets than individual stocks), then memorize and apply these three rules to your proposed wave counts.
2 Comments | add comment
Intraday Ascending Triangle Example in DIA
October 2nd, 2008 by Corey Rosenbloom
On Wednesday, the Dow Jones ETF - the DIA - gave us a good example of an ascending triangle… that brokeout in the opposite direction as expected. Let’s examine it and learn from its development.
5-Minute Chart DIA:

Generally, an Ascending Triangle is a continuation pattern and has an upwards break-out bias, as strength is thought to be building as evidenced by buyers stepping up to higher levels (higher swing lows) to support prices while overhead resistance remains constant.
However, this is not always the case, and triangles at their core represent a ‘pause’ or consolidation with an uncertain - though expected - ‘ejection’ or break-out point in either direction (though some triangles have built-in bias).
In this case we see the $108.60 level proving to be significant resistance, as the price repeatedly tests this level (even going back to September 30th) and price cannot break above this area - it is confirmed resistance.
The fact that higher swing lows are occurring sets up this consolidation pattern labled an “Ascending Triangle.”
We know from the price principle “Price Alternates Between Range Expansion and Contraction” to expect some sort of break, often close to the apex (convergence point) of the triangle and that is exactly what happened here - with this morning’s overnight gap.
In my experience, it’s better to wait for a price breakout instead of trying to enter inside the triangle. Why? In this case, the bias would have been to the upside and you would have entered long and then been trapped in this morning’s gap and downburst in prices and clearly would have been stopped out of the trade.
Stepping aside and waiting for the break, you would have sacrificed initial trade location in return for higher probability of a successful trade - entering short early in the morning. The overnight gap was roughly $1.00 in DIA terms, meaning there was a reduced probability of a successful morning gap fade trade.
When you see clean or clear patterns you recognize, print them, document them, annotate them, and store them in a notebook for further reference. The more times you see these patterns, the more able and confidence you will be to act upon them in real-time.
No Comments | add comment
Gap Fade Stats for September
October 1st, 2008 by Corey Rosenbloom
What a month September was for the Market. It lived up to its reputation as a weak month historically. So, how did the gap-fade tactic work in the Month of September? How many overnight gaps occurred in the Dow (DIA) and how many of those filled? It’s time to take a look as a new month is upon us.
I define an “overnight gap” as an open that is at least 20 cents in the DIA (20 Dow Points) different than yesterday’s (or Friday’s) close. A “Fill” is when price ‘gaps’ greater than 20 cents from yesterday’s close and then price equals or exceeds yesterday’s close. With that, let’s see the numbers:
A staggering and perhaps record 90% of the 21 trading days in September experienced an overnight gap of at least 20 cents.
Of these 19 gaps, 10 gaps ‘filled’ giving us a “Gap Fill” percentage of 52.63%.
Of the 9 “up” gaps, the average gap was $1.44 (144 Dow points roughly).
Of the 10 “down” gaps, the average gap was $1.41.
Roughly the same amount of up and down gaps filled, with a razor thin edge going to the up-gaps filling more than the down gaps (5 of 9 ‘up’ gaps filled while 5 of 10 down gaps filled).
What if the Gap Size was 50 cents (50 Dow Points)?
If we only classify a ‘gap’ as a 50 cent overnight price change (roughly 50 Dow Points), we return the following:
As goes the classic research, the larger the gap size, the lower the odds of a successful ‘fill.’ I find the same stats here.
Raising the gap to a 50 cent threshold finds 76% of days (16 of 21) showing an overnight gap and only 43.75% (7 of 16) gaps filling.
Just for fun (and one time only!) I thought I’d show you the magnitude of gaps that occurred in September. It may well go down as one of the record months in terms of overnight gaps (focusing on the DIA - the Dow Jones ETF):
A grand total of 1 day (9/9/08) had an exact match from yesterday’s open to the morning open and a second day had a change less than 10 cents. All other days gapped in some fashion.
We also had a $6.17 upside gap on 9/19/08 and a $3.64 downside gap on 9/15/08. There was also a $3.18 gap on 9/8/08.
Will October continue the trend? We’ll see!
To see prior months’ Gap-Fade Statistics, click on the corresponding month below:
January Gap Fade Statistics
February Gap Fade Statistics
March Gap Fade Statistics
April Gap Fade Statistics
May Gap Fade Statistics
June Gap Fade Statistics
July Gap Fade Statistics
August Gap Fade Statistics
Stay safe in these hyper-volatile times.
Update:
Rob Hanna of Quantifiable Edges recently released these posts on Gaps:
What The Recent Gap Action Is Suggesting About The Intermediate-Term
1% Gap Down Stats Since The Top
Gaps Stats And How The Lack Of Shorts Could Influence Action
Thanks to the NewsFlashr Business Blog Page for aggregating these links.
No Comments | add comment
Roller Coaster Market
October 1st, 2008 by Corey Rosenbloom
The only way I can describe the recent stock market activity is that of a wild “Roller Coaster” ride. Up, down sharply, up again, down again - nauseating, but fun if you like the wild swings (or have short-term strategies to take advantage of them). Let’s take a closer look at the S&P 500 to see some of these swings.
S&P 500 15-min chart:

We expected to begin the week on a high note with the passage of the “Bail-out/Recovery” Bill and I think Wall Street itself had priced in this legislative ‘victory.’ The Bill failed narrowly in the House of Representatives, sending the US Stock Market plunging along with the failure - giving us the largest one-day point drop in the Dow (777 points) and large percentage drops in all major equity indexes as well as a 10% plunge in Crude Oil prices on the fears of economic slowdown.
The technical structure at this time was that of a quite perverse bear flag that had vertical drops separated by a flat rectangle - forming one of the worst and most sudden bear flags I’ve ever seen.
Notice the 30 point swings in the S&P at the close of trading on Monday, September 29th. Up, down, up, down, close. Tuesday’s trading was a healthy ‘trend-day’ up which may have also defied expectations, especially if you followed the headlines “worst plunge ever” etc. With headlines like that, who needs friends? Seriously, the expectation was for a down market Tuesday but the Stock Market often has other plans in mind against the majority or even ‘common’ thought.
So here we are intraday on Wednesday and the market is - as of noon EST - falling. Down big Monday, up big Tuesday, down Wednesday. What a Roller Coaster!
What does the daily chart show?
S&P 500 Daily Chart:

The structure is, has been, and will still be a primary downtrend - price is making lower lows, lower highs, and is beneath all three key moving averages - and the averages themselves are in the most bearish orientation possible.
Caution is merited in this environment and the edge has shifted strongly to the professionals and away from the smaller/retail traders due to the increased volatity (and retail/newer traders not being accustomed to such large one-day swings followed by equal swings in the opposite direction the next day).
Again, capital preservation and risk-management are the top goals in this environment!
No Comments | add comment
« Previous Page — Next Page » |