Some points:
1 - In whatever timeframe, there's either going to be trend or chop going on. A quick glance at a chart in that timeframe pretty much gives you the lay of the land.
2 - Very long term, it's all chop right now: have a look at
the Privateer's monthly Dow chart, which pretty much defines secular. Since 2000 or so, it's all been chop.
3 - Chop in a really long term chart, like the Privateer's, is a very very bearish indication. If you see something like that, prepare for the worst, hope for the best. Sustained high volatility in a long term timeframe is not a good thing at all, and this is about as long term a timeframe as you can get and still stay within an investing lifetime. Unless you're trading with your grandkids in mind, this is as long as it gets.
4 - Getting back to practical timeframes, the way I do it is to have a way of defining a trend and going with it, while at the same time having a very adaptable and fast way of measuring momentum with the idea of catching a breakout to the opposite direction almost (never really simultaneously, but as close as practicable) as soon as it happens, which helps although it doesn't totally solve the problem of chop. Of course, filtering for noise, which is what widening your stops would address, is a reasonable adaptation to making sure you don't get out of a good trend before it's over. I use options rather than stops so that I'm always trading the same system with the same parameters, with the options hedging against sudden moves against me. I could never use stops correctly, partially because my timeframe is more than a day but not by much, so I'm always holding overnight, but trading pretty frequently. Stops just aren't practical for overnight gaps.
