Determining the trend

Let us define "trend" as: "the period of time from the first bar after a MA changes slope until the first bar after the MA changes slope again"

(ignore the rare instance when a MA is going sideways... depending on your settings, that might happen like for one bar during a week long period).

Any given time period has periods of MAs that will perfectly fit trends and periods of MAs that will chop you to death.

Unfortunately, they change constantly.

Say you're using 5 min charts. One day a 22 period MA might be perfect. The next day it might be a 37 period MA.

Say you're using 145 tick charts, or 4000 constant volume charts. One day 15 period MA might make you a million dollars, and then next day it would blow out your account.

Therefore using standard MAs, there is no way to consistently correctly determine profitable trend.

"higher time frames" are irrelevant because "higher time frame" is subjective. You can find higher time frames that fit the trend you want to see or higher time frames that don't fit it.
 
^ sorry I'm frustrated after backtesting tons of "trend following" systems all with different settings and none of them being profitable (unless during the testing time period you happen to catch a massive trend that wipes out all the losses).

Basically if you're a trend follower you are going to have a billion small losses and hopefully one absolutely incredible win to override all the losses and hopefully leave you with a profit.
 
Quote from IronFist:

Let us define "trend" as: "the period of time from the first bar after a MA changes slope until the first bar after the MA changes slope again"

(ignore the rare instance when a MA is going sideways... depending on your settings, that might happen like for one bar during a week long period).

Any given time period has periods of MAs that will perfectly fit trends and periods of MAs that will chop you to death.

Unfortunately, they change constantly.

Say you're using 5 min charts. One day a 22 period MA might be perfect. The next day it might be a 37 period MA.

Say you're using 145 tick charts, or 4000 constant volume charts. One day 15 period MA might make you a million dollars, and then next day it would blow out your account.

Therefore using standard MAs, there is no way to consistently correctly determine profitable trend.

"higher time frames" are irrelevant because "higher time frame" is subjective. You can find higher time frames that fit the trend you want to see or higher time frames that don't fit it.
More...

Ironfist,

Excellent post.

I've been through the MA strategy and that's why I keep coming to PA, whether you define it as watching the tape (DOM or T and S) or watching charts without indicators (though I haven't ruled out indicators yet.)

Anyway, the PA people define trend at successive HH and HL in and uptrend and the reverse in a down trend.

So using that definition and accepting that a higher time frame is subjective, then when does the trend start?

So far, it seems that one chooses the time-frame to trade in, for this thread, we're only talking about daytrading, then decides on a time frame for determining the trend and one for trading. OR just one time frame for both as Stealth Trader has said on occassion.
 
Quote from vingbel:

Iwhether you define it as watching the tape (DOM or T and S)
More...

I would love to learn those but a) my broker doesn't have TaS and b) I've spent days watching DOM and it's all random noise without any noticeable patterns or anything else (I guess I'm missing all the good info that everyone else on ET seems to be able to see). I've also never seen a single post, website, or book, explaining how to trade from DOM (plenty of them explain DOM, but none has any info about what to look for). Maybe it's one of those super secret don't talk about fight club type of things.

Oh well.
 
Its not just that easy to ''follow the trend mate''... lol

Because lots the time you see the trend the trend could have finshed and be changing.


I trade oil, and so i scale out from the 1minute all through to the 30mins-1hour chart.

My fav for entries are the 10/15minute chart.
 
Quote from spanish89:

Its not just that easy to ''follow the trend mate''... lol

Because lots the time you see the trend the trend could have finshed and be changing.


I trade oil, and so i scale out from the 1minute all through to the 30mins-1hour chart.

My fav for entries are the 10/15minute chart.
More...

Finally, someone speaking truth to power (or to ET boards!)

You got it.

1) First it depends on where you decide to look from. What's you point of reference.

2) Then you have to decide as it's unfolding. Is that a higher high or a reversal as part of consolidation?

3) But, assuming, you have a point of reference (even just the opening range, which, of course, itself has to be defined) then you still need to say what you think is trend -- And many seem to think that it's either 2 HH and 2HL or 3HH and 3HL.

4) So going with that, my question comes back to the reference point you choose.

And, I agree, even though their are threads that say, "hey just look for the 3 swing and enter as the retracement breaks," I'm not hearing that still works day to day for people. Even AHG doesn't seem to follow that anymore.
 
Quote from IronFist:

I would love to learn those but a) my broker doesn't have TaS and b) I've spent days watching DOM and it's all random noise without any noticeable patterns or anything else (I guess I'm missing all the good info that everyone else on ET seems to be able to see). I've also never seen a single post, website, or book, explaining how to trade from DOM (plenty of them explain DOM, but none has any info about what to look for). Maybe it's one of those super secret don't talk about fight club type of things.

Oh well.
More...

You can read the DOM and T and S, but it's matter of watching it for a long, long time and getting to know the trading vehicle you're trading.

There are others on these boards with much more experience than I who could probably list more pragmatic ways to learn and understand the DOM and T and S. But it will always involve sitting there and watching the stock/index/whatever go back and forth between extreme prices.
 
Cheers mate.. :)

The best way to identify how far through the current trend the market has moved, (since at the time when you are simply looking at a blank space on the chart for where the future will be its alot harder than we think, once we are now looking in hindsight).., the best way to identify the trend is to use just the basic S&R levels from the 15min/ 30min/ and 1hour chart, so you can see how many trend levels have been broken so far in the current direction the market is going, + how fast / steep that has been.


The faster and steep the fall, not always, but ususually means the trend could be coming to an end.

However like i found out yesterday in oil when the new day high or low is made this can also often start a new even further rally in that same direction.
 
So perhaps before you start trying to identify and trade the trend, which no matter what anyone says, this is extremely hard since the trend could start changing every 10mins / few hours for all you know.

So you could start and focus on working out the 'short/mid term trend based on S&R level peaks from the 15/30/1our + using news and fundamentals as a main key, and trading back against the trend with a tight stoploss, when you get big spikes.

This way you are limiting your risk of cash you can lose due to a tight s/l, and could also be getting in at the start of a new trend as a bonus.
 
You must log in or register to reply here.
Back
Top