rangerdoc

You're right, ADX is about the slowest thing out there. Personally, I don't use hardly any indicators anymore. I'll use one medium oscillator for divergances and one volume based indicator... that's it. And in their own window, never on the chart window. I keep the price naked. All these indicators are just derivatives of price. Now you can use the latest computer (quad core) that processes about 59,000 MIPS (Million instructions per second) to determine trends and price action... or you can use your brain which has a processing speed of 100,000,000 MIPS. Which do you think can do a better job of evaluating a price chart? I found that I was trading the indicators, not the price. Once I stripped all the indicators away, my profitability increased about 40%. This takes time though. Personally, it took me five years. Others maybe much more or much less. If that just won't work for you, then print the chart, hang it on the wall and call a five year old kid over and ask him if it's going up or down. If he can't tell, it's sideways. (In other words: use his 100million MIPS computer if your own isn't working) Kids have the huge advantage of not being biased in any way. Hmmm, Maybe I need to put a daycare on retainer???

What you're asking is really the million dollar question: If there was a way to reliably know when a market was transitioning from sideways to trending or vice-versa, we would all be very rich indeed. One can only really know after the transition has occurred. A problem here is defining a trend. After that, a "sideways market" is a lack of trend as you define a trend. There are about as many techniques to answer this dilemma as there are indicators. Multiple MA's, Comparing two different ATR periods, Rate of change in the ADX line or just looking at the chart. By it's very nature, you can not tell a trend until it is underway. Welles Wilder's ADX (sometimes called DMI) is specifically designed to differentiate between trending vs. sideways markets. Like any indicator, there are limitless ways to interpret it. Personally, I largely ignore the level of ADX (contrary to Wilder's methodology). What I use to determine trend is the direction and rate of change in the ADX line. Interpretation is largely subjective and very visual. Since it, like any indicator, is simply a manipulation of price, many would argue that you are better off visually detecting trend vs. range directly from price and not from an indicator, which will always lag behind price. Take a look into ADX. It's not a very intuitive indicator and therefore not very popular. It takes some getting used to. Those who are comfortable with it swear by it, and it often becomes their favorite indicator. Personally, ADX is the first indicator I want to attach to any chart and sometimes is the only indicator I use.

Not unless my psychology demands it. I find profitable trades to be more stressful than losing or stagnant ones. Probably because in those cases, I'm in control... I know exactly what I'm going to do. In a winning trade, you are completely at the mercy of the market. If I find the stress of a winning trade too great, I may scale out. It's purely to satisfy my psyche though and has no merit based on fact or research. Logically, if it is good to take a partial exit, then it is better to completely exit. Conversely, if it is a good thing to let part of a position run, then it is better to let it all run. IMO, in a scaling out strategy, you are always risking 100% of your position size, but only letting a portion of it run in a profit. This violates the golden rule. To me, it would only be logical if your methodology also calls for scaling out of losing positions also. Personally, my strategy uses a tightening stop. I reduce my trailing stop based upon how many ATRs of profit I have. A parabolic works well also. This way, my psyche is calmed by knowing that I'll be giving back less and less of my profits as the trade runs.