Interesting thread leovirgo, thanks for starting it. I have been having some fun with vwap bands on YM as well. I find they provide a useful point of reference and complement my PV analysis and setups.
If I am looking for countertrend trades then I would watch for price to "overextend" relative to a historical context (as opposed to just today's vanilla MPDs) such as yesterday's hi-lo / 2, daily ATR(x) / 2 or 15 minute ATR(x) *2.5, to name but three of an infinite choice of arbitrary partially-context-ignorant settings. Actually to my surprise I have found these three examples, using 10 period for ATR, which initially I chose nearly at random, to be quite useful across a range of market conditions, despite
this post. The benefits of historical volatility, or at least range, which I think ignores velocity, are starting to seep in just a wee bit. :cheesy: I use 24 hour YM not in-hours here.
Depending on the strength of the trend I have been experimenting with countertrend scalps as price touches or penetrates one or more of these bands. In fact I was feeling so nostalgic that I threw in a modified oscillator for confirmation. On a quiet day (e.g Dow +/-50) perhaps I'd take signals from the closest bands, but on a +/-100 day I'd ignore all but the outermost ones, to put it very crudely. To refine this approach perhaps I should look at the strength of the trend in a wider context so I am not fading the first band on a "quiet" day just after the hourly has exploded from a 13 hour base building exercise and is resting temporaily before another massive leg. Still, that's where discretion helps.
But I think it is fair to say that on any given day as price diverges an increasing % from trusty vwap, reversion becomes increasingly likely, or at least enough of it to grab 10-20 pts. So having more than one set of bands may be useful. Indeed I have found that doubling new scalp positions if further away bands are touched can be a helpful approach that helps pay for premature entries when oscillation turns into trend.
Occasionally price is very disobedient and flies straight through even outermost set of bands, but this is rare ... so far 20 points has been the worst since I started testing a few months ago ... lol not a long time I know
and I would always make sure I have enough bullets left to account for this and with luck would not have jumped in front of the train more than a couple of times. Even better, I'd be already on it, but I am a terrible trend trader.
I use PV criteria as well of course but find that I am benefitting from the definite entry point that the bands can provide.
The bands can also provide good scale out exit points if I am already postioned with the trend.
I would post some charts but my charting (Sierrachart) is a bit primitive in that the whole band is updated whenever, say, the 15 min ATR changes which means that one needs to follow them in real time to see them plotted correctly. I'm trying to fix this but my coding is lame.
I like your idea of using the MPDs to enter on pullbacks in a trend. At the moment I use different basic criteria such as ross hooks, or vwap itself / an ema after two minor countrend highs or lows (context willing) but the MPDs look promising. I think they might be good for using smaller size with a slightly looser stop than I generally like and a larger profit target, e.g 2/3 at vwap, 1/3 trail or opposite band, with initial stop at Grey's mk1 classic 2/3 of distance from entry to vwap. Plenty of testing to do first of course and I have to get over my impatience and hatred of stops wider than a few points.
But it would be nice to have an established small position of this nature around I could scalp larger size.
Hope I can add something more interesting to this thread in the future, or even start a journal. As you can probably tell the Fed has eaten all today's brain energy.
