Trend Lines Don't Work

What if people learn about some Economics and fundamental stuff along with placing orders at TL's. I had a short trade on EU and was nervous about it looking at those TL's , Fibs and S/R lining up against my small counter tend trade.

But i knew in the end it all depended upon what Trichet would say.
.

Well in the end no matter what the technicals say if the fundamental news changes dramatically then the market should go where the majority think it should. Doesn't mean right away and doesn't mean every time either.

That is why when I trade TL's, or anything else for that matter, there needs to be qualifiers to say yes or no and naturally a proper stop loss.

I think if you take a look at a trending pair on your charts, you can see that trend lines do work. Why? Psychology. Perception.

They also change, or fail. They are not sustainable. Nothing in the market is.

A TL is one of the most fundamental tools you can use. Saying they don't work is like saying S/R doesn't work. It's like saying TREND doesn't work.

Don't be mad at TL's just because every time you ever used one to enter a trade, it failed.

The question is, why do they work?

Do all professional big money traders say, "Let's buy when price reaches this line."?

Y'all are beating a dead horse. Scotty gave a great explanation on page 6 that /thread for me.

Imo, they work because they are easy to draw. When they are easy to draw, lots of people draw them. When lots of people draw them, lots of people will more then likely place orders around them. In my mind it isn't a whole lot more complicated then that. That being said, I wouldn't rely solely on a TL to make money. They are useful, but in themselves, more then likely won't make you $. I wish so, then we could all be rich. Just my opinion

if they dont work, then dont use them. problem solved.

the thread in a picture
http://www.cenekreport.com/storage/b...=1274642541111

great explanation scotty
marbles

I agreed with you. In his book "The New Science of Technical Analysis", Tom Demark commented that general technical analysis doesn’t work due to it being so subjective. Tom believes in simple, non-optimized, and objective mechanical systems, and believes they should be universal across all markets and all time frames and hold up under all bull market and bear market conditions.

That's why he created mechanical methods to draw trend line. Although it is not 100% accurate, at least, the result is consistent even if back test.

And another point Tom DeMark makes is sometimes an even more effective way to trade is to fade the expected outcome when the trading technique, in this case trendlines, fails.

In other words if an expected move up through a trendline doesn't fit his qualifiers then go short, and vice versa.

The best trades, which are usually the hardest to take, are the ones where a majority are fooled - which in this case is when the majority think a break has happened when it hasn't.

I hope I'm not too late to the party to have a whack at beating the dead horse lol! (great pic)

Back to the original question posted by the OP. I've thought about this question for a long time (it's really a shame more posters don't challenge the assumptions so props to the OP) and the more important (IMO) underlying question explored in my old thread "what is a trend?" http://www.forexfactory.com/showthre...=225112&page=1

One caveat regarding the thread (above): I believe I drew the wrong conclusions or perhaps didn't reach a conclusion in the thread. I'm not going to get into the long explanation here but let me just summarize my current thinking.

As a background I've coded up several versions of trend lines (automatic) and have tested them extensively (including Mouteki/Demark) but there is no edge there. The caveat is that I tested them on short term (intraday) charts. In this thread OldQuant and Ha-patterns (among others) both made some very good points.

The short answer to your question as to why trend lines don't work comes down to market structure IMO. On shorter timeframes the market is probably weak-random. This (probably) accounts for why my auto-trendline tests failed. On longer timeframes (say 200 day MA length) the market has an observable, measurable and exploitable trend component (lots of papers on the subject, do a search). Trend lines are a LINEAR tool. The markets (particularly on shorter timeframes) are non-linear, meaning you get twists and curves and it DOESN'T follow a straight line course and also isn't purely cyclical as the sine wave curve which makes life difficult for cycles followers. It is possible to come up with a static set of replicable rules (I have) and test them (I have) for profitability. The OP's question WASN'T "can you trade trend lines profitably?" I'm sure some have managed to do so but it is most likely by using them in conjunction with other tools due to the fact that in and of themselves there is no edge (that I've found).

By the way, Hi newyear, thanks for the indie im still using it!


OK if we take a step back and become a little...questioning


What does it mean that something works? Perhpaps we have an idea of what it means that something works. And perhaps it is this that must be clear before we can answer the question of if or if not something works.

I dont believe in coincidences. On the other hand i dont believe i know why things always turn out like they do, at least not exactly, can anyone say anything different?

Anyway i was fiddling around with drawing trendlines and chucking a few trades at the market while doing so and reading a few more or less interesting threads meanwhile. Picture tells what it looks like.

So lets wrestle with that:
How would i know if what is being done works? Is it not true that i must first know what works in order to know what works? I mean if im looking for something, normally i would want to know what it is im looking for, right? Otherwise, how would i know if i found it?

The easy way out would be to say that i DONT know if it works even though i have those lines and those numbers. But that might mean i will never know if my trading works, if i cant look at those numbers and those lines and tell if its working or not. So lets not go that route, lets assume, for some reason, we CAN now if it works. What would be the criteria? Now i think its getting interesting.

Can a stoploss of 6.000 pips work and an infinte amount of extra lots placed? No that dont seem very smart at all. Less then infinte amount of positions and less then 6.000 pips SL for each must be used.

Anyway, how we go about answering the question of if trendlines work, is a matter of the bigger context of what we are trying to do.

So all in all, i think there is no answer to a very general question of if they work or not. There is a pretty bad and not so satisfying way of answering the question: If you are using it 10 years from now it works. But why must the answer be any better then that? If i was trying to find the cure for some form of cancer, maybe id have to settle with an answer like that: I dont know if the method works that im trying, i guess i will have to see in 10 years when research grants have definitely dried up.

I think i often settle for answers like that. Its hard i know, i want nice good answers i can tell people at parties when they ask me what i do for a living, but im only human i cant get that good answers most of the time.

Maybe im just tired or something but maybe settling with vague kinda bad answers is good enough.



Anyway, i dont use trendlines normally, so if i did get something to work it wasnt that. Or i do use trendlines even though im not aware of it so thats why it worked. Or i do use trendlines and it didnt work thistime, it was something other that did work, though i dont know what that is. Or trendlines dont work, i drew them on afterwards to suit price nicely, even though i have memories from drawing them before it happend. Or im working for the banks spreading confusion so honest hardworking folks will get confused and give away their money. Or a mix of all of the above plus a couple of other things i nor anyone else will ever find out about.

Now my 13 hours workday is over i think so im gonna take a shower, catch ya later!

My trendlines always working.Works 80-90%, especially in H4 time frames. I used trendline zones... They are more valid.

Does trend lines work? Hey, good question NY . I never bother to ask why does prices reject at certain level, and what happens if it touch or break the lines? Knowing this extra information might give us better odds in trading.
What if someone knows in advance where and when the price is going to bounce/reverse and already placing his order beforehand?
I have a system developed by myself (manual trade), that can anticipate where the price is going to stall and reverse. Not everytime, but most of the time. I can pin point the exact price level (+/- 5 pips), but i still don't even know hows does this things happen. Or why and how it works. I trade blindly. But the good things is i made profits.
IF the majority of traders are having the same common goal and the predicting the exact same levels, then this level will be made true. This also known as self fullfilling prophecy. In my opinion, this theory holds true if this "majority traders" are having more accumulated money from the opposite side of the market.

The good things about trend line is, it can be use to predict the future value. Applicable to linear trend line.

in Forex I found that trendlines work better in higher timeframes,
in lower timeframes like 1 minute or 5 minute they don't work well,
there is a lot of fake brakeouts and violiting of the lines,
I think over 1 hour and above they seem more reliable.

What other kind of trend line is there?

Similar question to this, but does anyone know why Channel lines work?