In this issue, we'll take a look at a way of generating high probability turning points in the future using the popular Mirror Image Foldback tool. We've seen this technique once before in these pages, but it never hurts to review the good stuff.
The Mirror Image tool is really very simple - it just projects the past forward in reverse order, as if you were holding half the chart up to a mirror to see the other half. The mirror in our case is the vertical dashed line, which you can see on the chart below:
You can see that price followed this forecast until around March of 2004, when everything flipped upside down. That's OK, because the turning points given by the forecast are still valid. I've drawn lines and placed arrows on the chart at important turns in the Mirror forecast. All but one of the forecasted points corresponded with actual points in MSFT, so this was a good placement for our Mirror tool. Remember, all of this was known as soon as we placed the drawing tool at the low back in November of 2003.
This is a pretty good forecast, because the mirror location we selected was an important pivot. Not all pivots will produce forecasts as good as this one. Sometimes, a turn just doesn't materialize at all, as we saw in March of 2004, when the market fell through an expected low.
One method we can use to verify whether or not we can expect a potential turn to hold is by using other Mirror Images from different starting points. If a turn from one mirror is good, it should be even better if confirmed by a totally different mirror. There are plenty of important tops and bottoms on the chart above, so let's just go through, place a mirror on each one, and look to see where they line up.
To keep things from being too messy and hard to follow, I've only drawn in one mirror on the above chart, centered at point F. What we do is just look ahead for important turns in the mirror, then mark them off. So our gold mirror line rises strongly into late July, then makes a top and comes back down. That's one turn, so mark it with a gold arrow. Next, move the mirror back to point E, look for turns, and mark them. You just continue this way and eventually what you've done is created a composite forecast, based on all the important recent swings in the chart.
I did this on the chart above on points A-F, and have placed arrows at one particular area where the forecasts all came together. Since we're using Point F as part of the forecast, we don't look for any confirmations earlier than that. You can see that each of our 6 important locations has generated a forecasted turning point near the middle of July. Half of them say it's a bottom and half say it's a top, but we don't really care about direction - we just want the time of the turn, and the tighter the signals the better. In the chart above 5 out of the 6 arrows fall within a 3 trading day window, giving us a very nice target for a turning point.
As you can see, this one worked out well, with MSFT topping out on the middle day of our 3 day window. Once you have a zone mapped out, the rest is just using technical tools to get you in and out. Exhaustion Bars and Momentum divergence would have done the trick here.
It does take a bit of work to keep track of all the turns, but it's not a difficult process. Once you have the dates mapped out, you'll have some strong time zones to watch for turns. Remember, the best turns will happen when the mirrors line up tightly at one spot. If they don't, or the zone is wide (like a spread of 8 bars instead of 3 like we had here), then the odds really start to drop on the turn.
All for this installment. Happy Mirroring!
Back to Archives