Keeping it simple is always a plus: http://www.elitetrader.com/et/index...w-a-straight-line.287955/page-31#post-4074238
Exactly, no matter what time frame someone chooses, when the trend starts and when it ends will be randomly distributed. How does one develop a trading method based on those random occurrences? Perhaps do not use TA do generate buy/sell signals, and instead use future value models to determine if the current price is over/under valued. Of course we see how well that line of reasoning worked out for LTCM and AIG.
Got to be very long term for that, a days movement should be considered just noise and a 1000+ sl. The randomness is a pain, just got to get in early and exit when momo starts to weaken, ofcourse something reverse too fast buy hey!
Only define trend in the longest time frame that you look at. Everything on the lower time frames is analyzing where to enter and exit. Pretty simple and salient.
Thanks for the like Turveyd. I'm sure you know this, but SS's point about different trends in different timeframes is just pullback/retracement oscillations within the larger trend. If he wants to trade short against a longer uptrend, that's fine, but a shorter leash need be put on.
This is incorrect as it is not absolute, when a larger trend ends the exception makes your statement, invalid.
Buy dips or sell peaks, depending on what you think the trend is. When that doesn't work you probably have the trend figured wrong.
It seems by now we can all agree that there are several ways of defining a trend (successive bars being higher/lower, moving averages, slope of trend channels etc.). The gaping question now seems to be how to measure the likelihood of the trend continuing given that we have established that there is a trend.
It's tricky, I'd say a decreases rate of change, ie slowing down as it moves up, eventually being flat then who knows. BUT News events can reverse the trend and take all your profit away in 2mins so it's never that easy I'm afraid.