022803
Welcome to the first showcase page of 2003! It's been awhile since the last one, but during that time I've been working on a trading manual which can be thought of as one huge showcase article. I just sent it off to the printer the other day, so it should be ready sometime in March. There's no way I could have covered all the tools in Wave59 in just one manual, so I limited my scope to the tools found under the "Technicals" menu. If there's enough interest in this manual, I'll write another one about price and time.
Now let's turn our attention to markets and take a look at an interesting technique that I've been playing around with. Most traders are familiar with static time counts. You basically take an important high or low, then count forward so many bars to get potential turning points in the future. How far to go depends on what magic numbers you're using. So Gann traders might go in units of 45 bars, while Fibonacci traders might use the standard 8, 13, 21, 34, 55 cycle. Each of these counts represents a possible turning point so when you get multiple counts hitting on the same bar in the future you have an even stronger indication of a change in trend.
With this approach in mind, let's take a look at a recent chart of the e-mini:
This is a 3 minute chart of the last two days, and I've marked off three important turning points near the beginning. What we are going to do is use these turning points to generate future turns, but we'll do it in a unique way. Instead of using time counts, we're going to use mirror image foldbacks.
A mirror image foldback basically does what it sounds like. It projects what just happened into the future as if you had placed a mirror right on the chart. If you find a good inflection point, you'll see all the turns in the past being mirrored by corresponding turns in the future. A turn 50 bars in the past would result in a turn 50 bars into the future.
So let's say we're sitting here at the beginning of the 28th. We already know these three important turns have formed, and want to know what to expect in the future as far as new turns go. All we do is apply a mirror image foldback to each of these three turning points, then look for where the three turns all issue a signal at the same time. It's just like looking for static time count clusters, only we're doing it off a different tool.
Here's a chart with the foldbacks applied:
Once we've got our three mirrors up, we just need to examine them to see when we get turns in common. I've marked them off on the chart above with vertical pink lines. These places represent areas where three different mirror image cycles all say there should be a turn. To make sure it's clear, here's what the signals look like without the associated mirrors:
As you can see, these turning points were very accurate. One of the problems with trading mirror images is trying to find the right inflection point. Some turns give much better visual forecasts than others, and finding the correct positioning of the mirror can be a trial-and-error affair. But all important turns generate valid mirrors, and this technique demonstrates a very easy way to take advantage of this fact. We don't need to find the perfect mirror placement as long as we just use multiple mirrors to confirm and support each other.