Evolution of Timeframe

C

capitalMan

I remember reading somewhere that the markets is analogous to
a eco-system, (I think it was andrew lo) and there is
evolution going in the markets too...

I think that traders can be catagories by the
different timeframes in which they trade,
there are position traders, scalpers, swing traders just like
there are differnt species of animals..

Now...My opinion is that the only way to profit from the markets is
through following a trend in the timeframe you are trading...
and the

POPULATION SIZE

of the the traders operating in a specific timeframe
determines the profitability of the trader trading this timeframe...as the number of traders trading a timeframe increase,competition increase and the markets become
more EFFICIENT( yes.. EMH) , profits are arbitrage away,
and the markets becomes chaotic( I didn't say it was random)

but this process provides opputunity for other timeframe traders because crowd behavior and trending action is SHIFTED to different timeframe..

as this other timeframe becomes trending( and therefore profitable), everyone jumps on the bandwagon and profits are arbitrage away again and price action becomes choppy..

this process is evolutionary in nature and its dynamics is a funtion of relative population size of traders operating in a particular timeframe

the implications for the trader is that we should AVOID trading in a timeframe in which the majority of traders operate and where the population size is largest...

the job of a good trader would simply to trade trends in a timeframe that are "out of fashion" ..

this would probably explain why some people have trouble making money in the markets

EVEN THOUGH THEY ARE DOING ALL THE RIGHT THINGS,

( CUTTING LOSSES AND LETTING PROFIT RUN)

THEY ARE JUST DOING IT AT THE WRONG TIME...

There is simply not enough profits to go around in that timeframe.

(gekko...you really shouldn't be too hard on yourself...may be it's just the market)

I look around ET and nobody even mentions holding positions overnight anymore...everyone and their mom is a daytrader or scapler nowadays...may be it's time to re-consider or buy/sell and hold again?

I thought I wrote this down to clarify my thoughts b4 I loses it , does it make sense to anyone?...hope to get some comments on the subject
 
it's an interesting concept you bring up. kind of reminds of that traffic pattern thing aphie brought up a couple days ago.

is it based on anything more than anecdotal evidence or your own musings? not that there's anything wrong with that, you still might have hit the nail on the head. i'm just curious.

also, if it is indeed as you say, you might have been best served by keeping it to yourself. unless you couldn't develop the idea further and wanted assistance. in any case, i can't imagine there'd be too many people willing to drastically increase or decrease their trading timeframe. i know i wouldn't. but then what i'm doing is working, so no sense in stopping it.
 
yes, absolutely. Constant adjustment, not just time, everything. You think there is going to be some stable system out there? Always changing. I call it working, always always working.

Time frames, profits, stops, chops. Always changing. Sometimes they change so fast, the only way you can make money is to just not change at all.
 
Quote from vegasoul:

I remember reading somewhere that the markets is analogous to
a eco-system, (I think it was andrew lo) and there is
evolution going in the markets too...

(gekko...you really shouldn't be too hard on yourself...may be it's just the market)
More...
1. I think it was Niederhoffer, not Lo.
2. As far as excusing Gekko this way, I think that's by far the lamiest excuse I have read on this board.
 
Quote from vegasoul:

the implications for the trader is that we should AVOID trading in a timeframe in which the majority of traders operate and where the population size is largest...

the job of a good trader would simply to trade trends in a timeframe that are "out of fashion" ..
More...
i'm so glad you brought this up. this is something i think about, today. i'll sum up my point... it comes down to, you either want to see what the crowd sees, or, you believe you just need price movement--who cares why it's happening.

if you take the 2nd case, then you could use very uncommon timeframes. trade 7.2 min charts instead of a 5 min chart. if you trade a daily chart, you know there's a load of stops around the previous day's high and low. like i said, maybe you like this.

on the other hand, if you trade a unique timeframe, chances are your stops aren't where everyone else's are. you're on your own in the market.

i previous started a thread about this very concept:
http://www.elitetrader.com/vb/showthread.php?s=&threadid=8084

the general opinion was mixed, but maybe that's a good thing.
 
Quote from vegasoul:


I look around ET and nobody even mentions holding positions overnight anymore...everyone and their mom is a daytrader or scapler nowadays...may be it's time to re-consider or buy/sell and hold again?
More...

I do exclusively overnights... Why trade any other windows? that's where most of the action is (as long as it's not going against you :D)
 
Quote from vegasoul:
Now...My opinion is that the only way to profit from the markets is through following a trend in the timeframe you are trading...
More...
Not the only way. There are lots of short-term strategies (based on length of time in trade by bars, not time frame) that yield very consistent results that are not based on trends of any kind.


Quote from Gordon Gekko:
i'll sum up my point... it comes down to, you either want to see what the crowd sees, or, you believe you just need price movement--who cares why it's happening.

if you take the 2nd case, then you could use very uncommon timeframes. trade 7.2 min charts instead of a 5 min chart. if you trade a daily chart, you know there's a load of stops around the previous day's high and low. like i said, maybe you like this.
More...
This won't make much difference. Many traders mistaken believe they can turn a method with no edge into an edge by changing the time frames. The reality is any significant pivots in the market will show up on all similar time frames. It may be a 20-bar high on 7.2 charts but will be a 28-bar high on 5m charts, etc.

Also, if a method did work on a similar time frame but not on the other, then this is nothing more than an optimization that happened to be profitable, but does not represent a general approach that is successful. It is like having a 5/10 MA crossover that is unsuccessful and switching to 4/9 MA's and all of a sudden you find a winning system. What you should be doing is run a full optimization set on all similar indicators on all similar time frames and observe the 3D surface plot of profit vs all variables to see if you find a stable plateau which supports the approach you are using, or if you happen to arrive by chance at an unstable peak.

If you don't understand what the above paragraph means then I strongly suggest you resist researching new methods until you do.
 
Quote from dottom:


Not the only way. There are lots of short-term strategies (based on length of time in trade by bars, not time frame) that yield very consistent results that are not based on trends of any kind.



.
More...


you are right..Some of .the short-term strategies is base on fading the crowd, (overbrought, oversold)One of the examples that comes to mind is "turtle soup", which fades 20day high and low in the futures market..


This fits exactly with what I am talking about earlier

As the "turtle system" and Richard Dennis got so much publicity that everyone becomes a breakout trader,(population of trend traders increase) the markets becomes much more choppy in the DAILY timeframe but this CHOPPY action is seen as short term trends intraday...the trending action is SHIFTED down to a INTRADAY timeframe

"turtle follower " and other intermediate trend traders trigger such others stops order and this cause a lot of VOLATILITY intraday, with enough volatility , a lot of stragies become viable

I am probably incorrect in saying that the ONLY way to profits is trend following...but I think without VOLATILITY or if you like, price movement (again defined within a paticular timeframe), no one can profit from the markets.

As to question of whether this whole idea is just my philosophical musing or something I can prove...

the pieces of this notion has probably been in my subconsious for awhile but I haven't been able to put a finger on it,

it all came together in a moment of clarity this morning when I was taking a shit and without even washing my hands, I decided to write it down immediately...







:D
 
I think I pick some of this stuff up from chaos theory...
the notion of fractals, and self-similarity between timeframes, fat tails and distibution with fat tails...
and how the the Hurst component flutuate through time..

I think most traders equity curve reflects this idea too..you win a lot money then you give it all back...the fact that you win a lot of money means the kind of strategy you are trading (discretionary or systematic) profitable...and traders pick it up in their back-testing or just by looking at charts...everyone jumps on to the strategy and the edge lose its "sharpness" and you give it all back...

you are back to the drawing back and the process repeats itself again and again...then you discover money management, and realize you don't have to give it ALL back ...only a portion of it ...which is a funtion of your risk parameter...

the other guys who found your strategy is also losing money and after a period of choppy action..everyone excepts the die- hards like you declare that "trend-following no longer works" and move on to other more interesting strategy...of cuz ...at this very moment...the markets start to trend again...


I think this is why money managment is so important..
Longevity is the key...
 
Quote from dottom:

Also, if a method did work on a similar time frame but not on the other, then this is nothing more than an optimization that happened to be profitable, but does not represent a general approach that is successful. It is like having a 5/10 MA crossover that is unsuccessful and switching to 4/9 MA's and all of a sudden you find a winning system. What you should be doing is run a full optimization set on all similar indicators on all similar time frames and observe the 3D surface plot of profit vs all variables to see if you find a stable plateau which supports the approach you are using, or if you happen to arrive by chance at an unstable peak.

If you don't understand what the above paragraph means then I strongly suggest you resist researching new methods until you do.
More...
i know what you mean, but how would you go about doing something like that?
 
You must log in or register to reply here.
Back
Top