Ascending/Descending Triangle breakout bias?

Beginner Questions
1

In your experience, is there a bias for breakouts from ascending and descending triangles?

Babypips school says ascending triangles weakly favor downwards breakouts. But other sources say they strongly favor upwards breakouts. Contradiction :confused:

It seems an important factor to consider would be what direction the price was moving before the triangle formed, but no sources I’ve read ever mention that.

Below is EUR/USD 15M. An ascending triangle has formed after a long down trend. I wonder if it is more likely to breakout one way than the other. I also wonder if the odds would be different if the triangle had formed at after a long uptrend, rather than a downtrend.


2

Depends on what you are really looking at. Triangles can go either way imo. It just gives you a good breakout point. However that ascending triangle you have on your chart will most likely break to the down side. Reason for that is that is not an ascending triangle. From what I can tell on your chart (which is way to zoomed in for me to give a real opinion on) it appears to be a bear flag. It will probably show a better bear flag on the 1-4 hour. I just went ahead and took a screen shot of the bear flag to show you what I am talking about.

1022×497

I highlighted a previous bull flag (in blue) just to give you and idea. The one I think you are looking at is in yellow on a 4hr chart. I also highlighted where the bear run is most likely got in its sights. Those yellow circles are on a good number 1.300 and price has rejected off of it. I would say there are alot of orders hiding around that magic number. Keep in mind I dont trade the EUR/USD very often so dont take what I say as trade advise. But if I was to play that I would look for price to break the top of your triangle at LO to get people long since it looks like an ascending triangle. then it will shoot down stopping them all out. Final target would be around 1.300. I would usually tp at 1.3050 but in this case seeing that both attempts on my chart broke the 1.300 level I would just trail down my stop at 3050 to see if it runs those stops resting at the figure. If it does it could very well gain momentum and run further. I doubt you would see a run to much past it but why not give a 30 trailing stop to see what happens. Worse case you get stopped and give back 30 pips. You still walk out with 100 pips in the bank. If it does run deeper then that more in the bank. Again dont take what I say as trade advise as I rarely trade this pair because it moves when I cant trade. And if it moves when I cant trade then I obviously have not made enough (yet) to quit my day job.

3

I believe that you are right in taking into account the previous move and where the price is coming from. If it is coming off a strong base or existing move, the triangle is more likely to be a ‘resting area’ for the continuation move. Bob has some good examples in his picture.

If the triangle has a flat edge, like you have shown above, I have noticed that it is more likely to break the flat edge. Probably because supply or demand builds up at the flat edge level until all of the opposing orders are gone and price is free to break that level.

Just my observations…

4

Good call, guys, effectively predicting the market. By the way Bob, yup, that yellow shape you drew is what I was looking at.

I’ve posted a screen of what happened the next day and half or so below. The price didn’t really break out up or down right away, it kind of kept going sideways during the Asian session. But then broke out a bit upwards as you both predicted, kind of dallied around nearing the end of the Asian session, then shot down about 90 pips after the opening of the Europe session around 4am EST.

15 minute chart:

43.gif698×513 13.2 KB

After reflecting on the replies from both of you, it makes more sense that the direction bias is better determined by the surrounding support and resistance levels, and current long term trend.

Looking at the charts with hindsight, and input from you two, it’s clear that price was heading down, seems like this triangle was a brief consolidation, before the downtrend resumed.

Maybe it will test that support indicated with the yellow circle on bobmaninc’s chart.

On the 4H chart below, it looks more like a couple of inside bars to me. Which I guess, end up looking like triangles on shorter time frame charts.

4 hour chart:

42.gif698×513 12 KB

Or maybe it’s an ascending flag (a parallelogram). The blue lines:

45.gif698×513 11.8 KB

The green lines above, it looks like either an ascending wedge already formed, or an ascending flag (parallelogram) not quiet fully formed yet. Or maybe who cares what the heck we call the shape, circle, octagon, butterfly, gorilla, whatever.

A question for bobmaninc or anyone…

Just curious, why would the screen shot be a flag and not an ascending triangle?

Actually, I think you might mean a pennant or a symmetrical triangle, looking from your drawings. I think a flag is supposed to be a parallelogram, which is either slanting up or down.

Whatever the shape is, I’m wondering why you drew with a slanted top edge, rather than horizontal like mine. I’m not very sure which is the best way to draw these things.

5

Very good question. The reason I did that is a subjective one. If you look at your 15 minute chart you can see that S/R level just before price went into consolidation forming your triangle. I used that to base the triangle starting point. Reason is that was a logical level that price would retest before heading back down. If as you can see in your chart price did retest that level before shooting down. Like drawing trend lines it can be very subjective so I like to give a point to look to get in at. As you could see if you waited for just a break of the triangle you drew you would have been stopped out by going long. Big money knows what pattern was forming so pushing price to the up side would get the early birds long and easy targets. By using the resistance level found on a lower (or higher depending on the formation) timeframe. I now have a level of where the fakeout was most likely to head to and reverse. Hope that helps