Why I won't trade patterns

99% of my trading is pattern based. And the one constant I have come to accept is you must always wait for the confirmation.

Also, be aware of what kind of pattern it is. If it is a consolidation you can look to only take the trade if it breaks in the pre existing direction.

If you do not have a good feel for the pattern, take a pass.

If you think the move will be substantial and you HAVE to be involved bracket stops on either side and let the market dictate. But most cannot do this...it implies helplessness and confusion. IMO, we all are both of the latter in the markets grips. Go along for the ride, flow like a jellyfish with the tides....

Happy new year and good trading to you all!
 
Support 100% the comments by Mark and Inda.

Just like confluence is required in profitable trading, confluence seems to exist in those trading patterns correctly.

Anek
 
Mark, you said you don't take a trigger on a pattern just because it's a pattern. Likewise, I don't fade a move out of a pattern simply because it moved out of a pattern. As you said, interpreting price action is most important.

According to traditional technical analysis books, trading patterns entails buying a move above the resistance line and shorting a move below support line. Obviously that's over simplifying the method because you can't approach trading that way because all patterns are not created equal. I've never heard of waiting until the previous low is taken out before you open your short. To me, that is chasing a move, and also where the stops are. Once that low is taken out it sets up a double bottom with lower lows, and a trade I will take hard.

As noted earlier, it wouldn't surprise me to see someone who traded patterns successfully. The reason - and the only reason - I won't trade them is because my method is rooted in mathematics and mathematics suggest the win rate is very low. Now, I'll be the first to tell you that you can't use a hard science to do something that requires art. However, if the mathematics don't at least support the art then you are <i>probably</i> fighting a losing battle.

I know a lot, and I mean <B>A LOT</B> of people who trade patterns and almost none of them win. If you're winning on patterns then it simply means you're able to filter out bad ones from the good.

So you guys say you are winning with patterns, and I have reason to doubt you because you are probably that 5% who actually win at trading, which follows the traditional 5/95 win/loss ratio. It just so happens that you are trading patterns. They just aren't for me.

This post was intended more for the 95% who aren't winning and to suggest that perhaps they'd have a better shot at achieving success if they just <i>tried another way.</i>

Quote from NihabaAshi:

You've described a perfect example of why I will continue saying to other traders that just because you see a pattern...

That doesn't imply it's a trade signal.

Simply, your chart represent a simple pattern but it does not represent a trade signal.

Thus, pennants are not trade signals all by themselves and that's the reason why traders as you will struggle with them.

Therefore, a pattern should only be used to alert you to be prepared to trade because a trade signal may appear.

In addition, a price action does not merit a trade unless you understand the price action prior to the appearance of any patterns or trade signals.

My point...there is distinct difference between patterns and trade signals.

For example, that pennant pattern you show...

The actual trade signal involves the trendline break.

Therefore, the main issue here are trendline breaks.

Are trendline breaks to the downside reliable after a Higher Low price action in comparison to trendline breaks to the downside after a Lower Low price action???

Are trendline breaks to the downside reliable when the breakout occurs at/near a key support level???

My point is that the pennant itself is not your problem.

Your problem is that you need to learn more about trendline breakouts and support/resistance levels or zones.

There's a lot of stuff occurring on your chart example and the pennant was just a clue or warning to be prepared to trade...

Profitable traders saw more than just a pennant and trendline break.

Long or Short depending upon your trade signal.

That's the reason why different traders will trade your pennant example differently...

Mainly due to the fact that they are using a different trade signal.

One guy goes Short and another guy goes Long upon seeing that trendline break below the pennant.

Same pennant...different trade signal.

Successful trading involves many chapters of the book called Profitable Trading.

There's one chapter in that book called Patterns and another chapter in the same book called Trade Signals (a.k.a. entry signals).



Your next statement above implies you actually do trade patterns.

In fact, you stating that you fade patterns.

Just keep in mind via using your example...

You are now implying you would have faded that trendline break to the downside via going Long instead of Short.

Essentially, the pennant was your alert and the fading of the trendline break is your trade signal although I'm sure there's other things involved in that trade decision.

Regardless, you do use patterns as part of your trade methodology via fading them. :cool:

Mark
More...
 
So you guys say you are winning with patterns, and I have reason to doubt you because you are probably that 5% who ---------->>>> this should have said, "...and I have <b>no</b> reason to doubt you......."

Quote from ProfitTakgFool:

Mark, you said you don't take a trigger on a pattern just because it's a pattern. Likewise, I don't fade a move out of a pattern simply because it moved out of a pattern. As you said, interpreting price action is most important.

According to traditional technical analysis books, trading patterns entails buying a move above the resistance line and shorting a move below support line. Obviously that's over simplifying the method because you can't approach trading that way because all patterns are not created equal. I've never heard of waiting until the previous low is taken out before you open your short. To me, that is chasing a move, and also where the stops are. Once that low is taken out it sets up a double bottom with lower lows, and a trade I will take hard.

As noted earlier, it wouldn't surprise me to see someone who traded patterns successfully. The reason - and the only reason - I won't trade them is because my method is rooted in mathematics and mathematics suggest the win rate is very low. Now, I'll be the first to tell you that you can't use a hard science to do something that requires art. However, if the mathematics don't at least support the art then you are <i>probably</i> fighting a losing battle.

I know a lot, and I mean <B>A LOT</B> of people who trade patterns and almost none of them win. If you're winning on patterns then it simply means you're able to filter out bad ones from the good.

So you guys say you are winning with patterns, and I have reason to doubt you because you are probably that 5% who actually win at trading, which follows the traditional 5/95 win/loss ratio. It just so happens that you are trading patterns. They just aren't for me.

This post was intended more for the 95% who aren't winning and to suggest that perhaps they'd have a better shot at achieving success if they just <i>tried another way.</i>
More...
 
Quote from ProfitTakgFool:

So you guys say you are winning with patterns, and I have reason to doubt you because you are probably that 5% who ---------->>>> this should have said, "...and I have <b>no</b> reason to doubt you......."
More...

Hey PTF

You may be composing a reply to my earlier question right this minute, but I just wanted to bug you one more time about this because I really want to understand what you're saying; it is matching up with something I have been noticing in the markets intraday, but it's not something I have been trading (yet).

http://www.elitetrader.com/vb/showthread.php?s=&postid=1729233#post1729233

http://www.elitetrader.com/vb/showthread.php?s=&postid=1729237#post1729237

No need to read them I guess, just read below if you want.

(EDIT: This is a bit redundant now that I know you've read the earlier posts but I'll leave it up just in case anyone else finds it on this page of replies)

My latest summary of your approach (I realize now when you posted that link to Anek which said 'This is the trigger - a short on a break of support', you didn't mean that you were shorting, you meant that pattern guys are shorting that break, right?).

1. Prices through the day are randomly distributed (The F Distribution, although I am just inferring what that means, I have almost no sadistics beyond normal distro and understanding fat tails and mean reversion).

2. Occasionally, there occur pockets of normally distributed price events. These can (sometimes? always?) be described as the 'patterns' under discussion.

3. When prices break out of these patterns, that represents an event in the tail of the normal distribution. These moves are faded in anticipation of a reversion to the mean. A trade is not always entered - the art of it comes in deciding which breakouts to fade.

Is that accurate?

If this is accurate, what will be really interesting is whether you always wait for the pocket of normally distributed price events. After all, sudden, long, unidirectional moves occur without having 'broken out' from one of the traditional patterns. Do you ever choose to fade these? I know a lot of guys are trading by fading these moves almost exclusively, a technique which requires a lot of patience and the willingness to be wrong small and re-enter.
 
These are some very good questions but it will take some time for me to answer them. I'll get back to you either later today or tomorrow. I'll see if I can dig up and example for you as well.


Quote from traderNik:

I wonder if you'd mind giving me a quick clarification here. I understand that you're fading moves which represent events out at the tails of the normal distribution.

1. So do you have a way of quantifying the actual values you are looking for (some sort of real-time scanner that you can start to monitor when you figure you're in the normal distro part of the day)?

Or is it that the patterning itself is evidence of the normal distribution forming up? Hmmm... maybe this is it. Aha....

2. I'm not quite sure I get how you're finding the numbers that go in the tails in the moves coming out of a pattern. So you mean that regardless of which way prices tend to break out of the pattern (the pattern, which is evidence of the intraday formation of a normal distribution of price) you tend to fade it?

As I'm typing this I think I'm figuring it out a bit better. Is this an accurate summary?

Prices move randomly for most of the day, with isolated pockets of normally distributed values. 'Patterns' can (do? sometimes do? are the only things that do?) indicate these areas of normal distribution. Large moves out of these patterns are faded. The entry point is determined by...

(a statistical value, i.e. deviation from intraday mean? that combined with approaches to S & R?)

Thanks in advance, hope this isn't asking for too much detail. As you can tell, I did not take 'sadistics' in Uni.
More...
 
Quote from ProfitTakgFool:

These are some very good questions but it will take some time for me to answer them. I'll get back to you either later today or tomorrow. I'll see if I can dig up and example for you as well.
More...

Excellent, and thanks. I appreciate it. I'll have my head in the charts until Sunday night. I'll check back occasionally, take your time.

Nik
 
Like most posters here, the original poster is really saying "I don't know how to do this, and I am hoping that someone who does will simply offer the solution"

Okay, well I may not be willing to put a turnkey solution in print, but I will point a finger in the right direction for you.

First, the moving averages do offer some added value. Unfortunately for this market you have chosen the "wrong" ones. Put both 80 period and 3 period exponential MA's in place on your chart. Offset the 3 period by 2 points positive. This is step one.

Step two, you need to find a way to dynamically update your chart to show the effect of volume as price hits specific price areas. Also you should know at any given time where price stands in relation to the daily VWAP. The reason for this is that both institutional desks and program traders are using it or some variant of it during the day. On both counts I am going to remain mute. There are a couple of possibilities including volume oriented studies like price at volume and "market profile" (trademarked)

Step three, in case you do not have success with the preceding, I suggest you include pivots. Although they are static thus limited in their use, they do have a following and so offer some value for short term (intraday) trading. If you are going to use them, be sure to include daily, weekly and monthly on your chart.

Do this and re-publish your chart.

Oh and by the way, the moving averages are estimates. In point of fact there are no "magic" or single moving averages that offer special value to your trading. What you are doing is using the offset between the long and short in a specific way. Frankly I don't want to get into it, but IF you DO have a background in statistics, in theory you would have already known this and had a suitable setup in place. lol

It will be interesting to see what you (all) do after this.

Steve
 
Quote from steve46:

Like most posters here, the original poster is really saying "I don't know how to do this, and I am hoping that someone who does will simply offer the solution"
More...

I know exactly what you mean - it's clear in most cases like this that the OP either secretly wants help or more likely has tried and failed at trading a particular setup and is now running it down in order to convince himself that he's not a failure.

In this case, however, I'll have to respectfully disagree. I don't get that vibe from PTF at all. In a few different threads he has been pretty clear that he purposely avoids trading the typical patterns and wouldn't do it even if you showed him that you do it profitably. In fact in another thread he explicitly stated that he isn't saying it's impossible to profitably trade patterns, just that he doesn't do it and, as an editorial comment, he doesn't believe that too many people can do it.

How many guys here lie about their results? 70% 80% More? Probably more. So we can't know if he's telling the truth. I am willing to check out his techniques - especially if they can be codified statistically (makes them a hell of a lot easier to test that 'price has broken support', which I also trade).

Anyhow, difference of opinion is what makes a market.
 
Steve, you have very little credibility to be making a statement like this. I am not on a quest, asking someone to teach me how to trade. If you take the time to visit this thread you will clearly see, I have <i>figured it out.</i>

http://www.elitetrader.com/vb/showthread.php?s=&threadid=112126

Quote from steve46:

Like most posters here, the original poster is really saying "I don't know how to do this, and I am hoping that someone who does will simply offer the solution"

Okay, well I may not be willing to put a turnkey solution in print, but I will point a finger in the right direction for you.

First, the moving averages do offer some added value. Unfortunately for this market you have chosen the "wrong" ones. Put both 80 period and 3 period exponential MA's in place on your chart. Offset the 3 period by 2 points positive. This is step one.

Step two, you need to find a way to dynamically update your chart to show the effect of volume as price hits specific price areas. Also you should know at any given time where price stands in relation to the daily VWAP. The reason for this is that both institutional desks and program traders are using it or some variant of it during the day. On both counts I am going to remain mute. There are a couple of possibilities including volume oriented studies like price at volume and "market profile" (trademarked)

Step three, in case you do not have success with the preceding, I suggest you include pivots. Although they are static thus limited in their use, they do have a following and so offer some value for short term (intraday) trading. If you are going to use them, be sure to include daily, weekly and monthly on your chart.

Do this and re-publish your chart.

Oh and by the way, the moving averages are estimates. In point of fact there are no "magic" or single moving averages that offer special value to your trading. What you are doing is using the offset between the long and short in a specific way. Frankly I don't want to get into it, but IF you DO have a background in statistics, in theory you would have already known this and had a suitable setup in place. lol

It will be interesting to see what you (all) do after this.

Steve
More...
 
You must log in or register to reply here.
Back
Top