The Trend

[Evan]

[ QUOTE ]
But this type of trading is more of an artform than a science.

[/ QUOTE ]
This sounds a lot like "I don't really know what I'm doing and there's a good chance I'm running hot" to me.

[ QUOTE ]
There is no magic formula.

[/ QUOTE ]
There's no magic formula for playing limi hold'em either, but it's certainly more science than art. Basically, this sentence is useless and irrelevant unless you actually think someone here needed you to confirm that you were no actually trading based on magic.

[mrbaseball]

[ QUOTE ]
There's no magic formula for playing limi hold'em either, but it's certainly more science than art

[/ QUOTE ]

What about no limit? Certain areas of trading are certainly scientific and math based. Chart reading and technical analysis are not. They are all feel and experience and trying to create some sort of black box is an excersize in futility.

I've stood in trading pits or have been staring into a trading screen for close to 20 years and if there was a mathematical solution the business I am in wouldn't exist.

[eastbay]

[ QUOTE ]
[ QUOTE ]
But this type of trading is more of an artform than a science.

[/ QUOTE ]
This sounds a lot like "I don't really know what I'm doing and there's a good chance I'm running hot" to me.

[ QUOTE ]
There is no magic formula.

[/ QUOTE ]
There's no magic formula for playing limi hold'em either, but it's certainly more science than art. Basically, this sentence is useless and irrelevant unless you actually think someone here needed you to confirm that you were no actually trading based on magic.

[/ QUOTE ]

No, there is no "magic" formula to limit hold 'em but there is most certainly a way to quantify an unbeatable strategy. It just happens to be extremely complex, however, in principle it can be computed, and in practice very good approximations to it can be computed.

However, that's neither here nor there. We're talking about trading now, and of course there is no "magic." I am simply asking what people are talking about when they speak of "trends" and "following the trend" to see if there is any merit in the idea.

eastbay

[eastbay]

[ QUOTE ]
[ QUOTE ]
What is the actual criteria that separates uptrend from non-uptrend?


[/ QUOTE ]

It depends? What is your timeframe?


[/ QUOTE ]

My timeframe? I'm asking you guys what a trend is. I'm trying to make as few suppositions as possible so as not to prejudice the discussion.

[ QUOTE ]

If you want to test things (which it seems) I find moving averages as the best trend identifiers. Use a crossover or a longer term MA (ie 200 bars). If you are above it,it's an uptrend and below it a downtrend.


[/ QUOTE ]

Ok, there's an operational definition of a trend: if the current price is above a 200 bar average it's an uptrend.

Ok, that's actually a fairly "good" definition in the sense that I'm looking for.

Now onto part two of the question: How does one "follow" a trend once it can be identified?

eastbay

[mrbaseball]

[ QUOTE ]
What is "the trend" and how does one "follow" it?

[/ QUOTE ]

Sniper's response was great. This is how I use it and trade with it. I look at 3 charts (timeframes) of whatever I am trading. Usually 5 minute, 30 minute, and 60 minute.

I also use a moving average crossover. If the fast average is above the slow average on all 3 timeframes I consider it an uptrend and if fast is a below on all 3 a downtrend. I then make a trend following trade when the price rallies or dips to the slow average (on shortest chart usually 5 minutes). Obviously I use other things as well, such as trend lines and support/resistance for trade entries. But when I have met the condition of a trend on all 3 time frames and the price moves to the slow average on the short timeframe chart I am looking to enter the market in the direction of the trend.

[Officer Farva]

Here is the best piece of knowledge I have about markets and trends:

use as much relative information as possible, and value them all roughly equally.

for me, this means using a large set of quant factors (rsi, macd, ad, ma cross, 2nd derivatives, mean reversion, unit root tests, etc)

I then take some sort of fundamental idea and dump it in to this framework, and do it the same way everytime.

The key here is that although noone uses all of these factors everytime, there is always someone using each of these factors. So if you can get them to roughly be aligned, chances are there are more people in alignment with you. Combine that with a decent idea producer, and you have yourself a market beating trend model. I learned that first hand from one of the top macro managers in the world.