that should be the case with any system development. It is not the specifics that can be applied throughout the different assets or symbols but rather the general ideas. In any system, there is inherent bias in anything we do. For example if you happen to start researching a system in summer where vol is low, you are already biased in that such environment.
In reading your comments and questions a glaring omission became painfully clear (this is probably a bit of an overstatement) which I would like to correct at this time—which was my failure to mention that my description then (and what will now follow) is not, technically speaking, precisely accurate language. But first, let me admit that I cannot answer your question on ticks. I am not clear on exactly what a tick is. Every time I read up on the topic, the definition of the term fades from memory soon after. Moreover, when I study tick charts, I feel what I am looking at does not match (accurately reflect) what is going on in reality. The result of such episodes is that my motivation to look into ticks more deeply is nonexistent. You commented that “in time based charts, slope is (y2-y1)/(X2-X1) where it is essentially the change of price vs. time.” This reflects my understanding of slope, but when it came to representing this on MT4 candlestick charts, I was unable to devise a valid means of doing so. I thought perhaps a rate of change (ROC) indicator might do the trick, but in trying to apply its use, I found that it did not turn out to be practical for me personally. The closest I could get to something anywhere near what I might view as functional was to divide a moving average in its present state with where it was at some time in the past. Most recently, I was choosing to divide it by where it was one period ago. The resulting indicator (oscillator) was above zero when the moving average was headed north, and below zero if the moving average was headed south. Unfortunately, this too I did not find to be of great practical use. With respect to your assumption that one of the moving averages is short-term and the other is long-term, this would depend on how you define these two expressions. In one sense, both of the moving averages I use to dictate when I enter and exit positions are short-term moving averages. The longer/slower of the two is the one I regard as tracking the intraday trend with the greatest amount of precision, validity, and accuracy possible without subjecting a trader to head fakes or false positives. The shorter/quicker of the two conveys the shortest-term trend by measuring price direction in the smallest possible meaningful units, and is therefore extremely accurate, but also subject to frequent fluctuations, so cannot really be trusted as an “actionable” indicator. What I do is compare the faster indicator to the slower one, but the slower one is actually pretty fast itself. So then, I am not really measuring the slope of either trendline. In the language you used in your initial post, I am measuring the difference of the two SMA periods (by dividing the faster moving average by the slower one). Because they are both accurate measures of short-term trend, when they are moving in tandem with one another, it leaves little doubt as to which direction price is headed from an intraday point of view. One might think of them as confirming each other, and when this is the case, the momentum in the given direction is quite significant. When the faster indicator begins distancing itself from the "slower" one (relatively speaking) there is still no question about which direction price is headed, but momentum is increasingly extreme and on its way to becoming unsustainable. Technically, your comment about the short-term slope being flat or inverse of the long-term slope before the two cross over is correct, but because the two moving averages I use are not that far apart, this is not what usually happens. More common is for the two of them to reverse direction almost simultaneously, as represented in the image I posted where the two of them were trending south before reversing direction to trend north. The second most common occurrence is for the fast line to pull away from the “slow” line in the same direction of the intraday trend or opposite the direction of the intraday trend, then reverse direction to move back to the “slow” line, and reverse direction again (or decreased its radical slope) to more-or-less rejoin the “slow” line in its general trajectory, whichever direction that might happen to be.
I agree with your general statement. However, the title of this topic is about using MA for determining momentum of an instrument. Some instruments move much faster than other instruments. This makes that, if you find a pair of MA's which work well for one instrument, probably don't work for an other instrument, which moves much slower. @expiated had not clarified that his pair of MA's are only to be used for forex trading, which is why I wrote my response to his message.
I agree with HobbyTrading 100%. The (two) primary moving averages I use when trading foreign currency pairs are completely different from the (two) primary moving average I use when trading U.S. indices. I specifically stated that I use them exclusively on one-minute charts. However, because I was primarily commenting on the general idea (i.e., the concept) and not trying to offer specific advice, it seemed less pertinent to me to mention that I also apply them exclusively to the Forex market, but perhaps I thought wrong.
I was pleased with the potential that using the difference of two SMA periods as a gauge of momentum seemed to hold when I began exploring this tactic on Friday. However, on Sunday I noticed that sometimes when an instrument appeared to be gaining momentum, it would be quickly lost by a pullback. In seeking an effective means of dealing with this potential challenge, I made two additional observations: The first was that the pullbacks in the market I trade (Forex) are more-or-less cyclical and of equal duration. Second and of significant importance was that increases in momentum appeared to be accompanied by reductions in the size of pullbacks that were inversely/directly proportional to the strength of the momentum. In light of the above (three) observations, it occurred to me that if I took the duration of a typical pullback (let’s use a hypothetical time interval of one hour) and compared the actual distance covered during that period to the amount of progress made in twice that span of time (in this hypothetical—two hours) it would yield a more accurate measure of momentum than comparing two moving averages. The way I did this was by taking the average of the sum of the distances covered separately during these two different time intervals. If pullbacks were eating into the momentum, then the rate of change for two hours (using the above hypothetical) would be slower than the rate of change for one hour. And in fact, if pullbacks were significant enough, the amount of distance covered in two hours might actually be less than the amount of distance covered in one hour. This would of course be reflected in the resulting chart... On the other hand, a reduction in the size of the pullbacks would translate into steady progress, in which case, the momentum oscillator would maintain a path restricted to the light green or light pink areas of the lower panel graphic, which represent significant bullish and bearish momentum respectively (as illustrated by the situation depicted all the way to the right of the above image). Should an asset began to experience “monster” momentum, the oscillator would shift to the darker areas at the top or bottom of the lower panel and remain there fore some time. This too can be observed in the above image. (Of course, lack of momentum is conveyed by the white central area in the middle, which I might just widen a tad bit based on what I see in this particular chart, but more analysis will be necessary before making a final decision.)
It will probably be nice to have momentum represented in the lower panel, but being as visual as I am, I would be even happier if I could see it displayed in the main chart as well. I thought that perhaps shifting the line graph might work, but I did that in the past and was not happy with the looks of it. Then after a bit of analysis, it seemed to me that turning to the difference between the bands of two SMA Envelopes associated with key moving averages and set at the right deviation levels just might work, so I tried it out and liked what I saw... I plan to enter long position when price enters or climbs above the green zone, and enter short positions when price enters or crawls below the pink zone, though deciding where to exit with profit might be a bit trickier. (No doubt, I will add moving averages to aid that process.) It will be interesting to me to see if these new tools actually turn out to be helpful during real-life trading once the holidays are over.
@expiated sorry, but now you got me confused. The chart with the green and pink bands, combined with your description of entry points, give me the impression that you are identifying rising/falling trends and respond to those. It seems more like a trend-following than a momentum system to me.
The original post asked if anyone uses the difference between two SMA periods as a gauge of momentum. However, in looking into the difference between trading momentum and trading trend, I suppose that technically speaking, simple moving averages would, by definition, have little to do with momentum trading. From what I read, momentum trading is concerned with the strength of supply and demand inputs based on earnings reports for publicly-traded companies, the relationship between buyers and sellers in the market, the typical rate of historical price rises and falls, etc. In short, momentum trading is primarily concerned with the fundamentals. However, in response to the question about simple moving averages, I conceptualized momentum as the amount of strength compelling price in the direction it is trending. Given this informal “lay” definition (as inaccurate as it might be) I would avoid entering trades during periods when the formation of candlesticks is relegated to the white strip between the shaded areas due to a “lack of momentum.”
This is pretty much how the lower panel and main chart will look after being combined, though I just noticed I need to broaden the shaded bands in the lower panel to match the shaded areas in the main chart.