Thanks for the funny posts! A few serious answers, one half serious, the others had me in tears. I suppose I should now ask, if for example the 20 day MA is based on the trading days in a month (remember it's trailing), then how is a month's worth of trading days significant? It could well be what a couple of you suggested (about an accumulated following of a rule), or it could do with how we gravitate towards whole figures; weeks, months, years. A bit like how a stock sometimes stops on the multiples of 10's, or 100's. What do you think,?
Different MA lengths pick up different length trends. Trends occur at different time periods. There is no significant difference (in the statistical sense) between the pre-cost performance of different length moving averages in the period when financial assets are generally held to trend (say between a couple of weeks and a year or so, give or take). Ratios of 1:2 to 1:4 between the fast and slow moving average crossovers work best at capturing trends. This is true with artifical and real data. Given a piece of paper, a pencil, and some stochastic calculus I could probably show why this is the case and find the 'correct' ratio. This about as much science as you will get for this question. It matters not, since there is no significant difference between the performance of ratios in any sensible range. Do not, as someone has said, try and optimise the moving average to use unless you do so in a robust way. Better to use many moving averages, or crossovers, and take an average. It's intuitive to use logical MA lengths. Using business day calendars, 10 = two weeks. 20 = about a month. 40 = about two months. 60 = about 3 months, 80 = about 4 months, 120 = about 6 months, 160 = about 8 months. Doubling the MA each time gives a similar correlation pattern. I use 10,20,40,80,160,320. Or try working downwards from a year in base 2: 256 = about a year, 128 = about 6 months, 64 = about 3 months, 32 = about 6 weeks, 16 = about 3 weeks, 8 = just under 2 weeks. None of this will make you extra money, but it will discourage you from trying to fit 31 or 29 day MA lengths. GAT PS Some pics This shows, for Eurodollar, the Sharpe Ratio (Z-axis) for crossovers A and B. Any trend following system of medium length (lower left triangle) works pretty well (bright yellow). This shows the T-statistics (bootstrapped, non parametric) comparing the optimum to the relevant system. Anything in dark blue can't be distinguished from the optimium, and is just as good as it (optimium is white). There are a wide range of possible crossover pair values in dark blue. Finally this shows the T-statistics with all insignificant results whited out. Any trading system in the white area is as good as any other. So for example 30,175 is as good as 8,20. Another way of looking at this, here is the box and whiskers plot for some commonly used crossovers (averages taken across 40 futures, pre-cost returns): Apart from the first two, there are no significant differences in performance.
Moving averages are pretty useful. Understanding them and how they are used is also important. Really, you need to understand the convergence properties of families of moving averages. For example, take a few of them, simple, doubly exponential, wilder's smoothed, or a recursive median filtration. The first is an exponentially weighted moving average (EWMA). It calculates an average that favors the most recent price data using a variation on exponential weighting. (fast adapting) The second is a simple average. It assigns the same weight to all values in sample. (simple detrending) The third is wilder's. It's like a EWMA but takes a smaller sample. (smoothed and fast adapting) The last one, a recursive median filtration, or RMF. This one filter's the data using relative volatility, and so it reduces the weight of data in sample that is the result of rapid volatility changes. (robust detrending) You can chart a bunch of these things using overlapping periods to create estimates of trend and volatility changes. (momentum) This is a suite of filtering tools for price processes. You should know which one is used for what. An example would be using a bunch of simple MA's to do MACD. There are also regression families of moving averages. Polynomial regression, linear. This stuff is used to denoise, detrend, and obtain relative measures of price process data. Quants are using tools like this to build extremely sophisticated models that can be adapted to the sums, indexes, baskets, and spreads of financial instrument transaction data.
The figure 20 has not significance over say 18 or 23, but humans like round numbers. Like, 5 fingers per hand, 10 fingers total. Its why round numbers have alternative names like dozen, score, pony, grand, monkey, fortnight, century and so on.
It's not the indicator itself that matters but WHO monitors that indicator. If it's average Joe Daytrader, then it's not important. But if it's Steve Cohen, for example, then you better pay attention. In the same refrain, 20-, 50- and 200-day MAs are only important because they're being followed by the big money, eg. mutual funds, banks, etc.
Moving averages are lowpass filters. The most important parameter for trading is the filter's group delay, otherwise known as lag. To be useful for trading the group delay needs to be zero or almost zero. Also, to be useful, filters need to be causal (you can look up that definition.) Next in importance comes the amount of attenuation in the stop band frequencies of the filter. You want as much attenuation as possible while still trying to maintain zero group delay. Your filter's cutoff frequency controls how much attenuation and group delay you will get. Physics says you can't have a causal, zero group delay filter, but you can get damn close. Typically, the attenuation will be terrible however. My recommendation is to first look at John Ehlers' Zero-Lag filter, and then look into Kalman filtering.
So you believe that MA's work because enough people to believe them that in combination they buy or sell sufficient to move price? So indirectly, MA's move price? Are you really saying these things?
Long term moving averages are used by Wall Street. Finance professionals may use them to justify investment decisions.
Where price is in relation to the 200 is without question a factor in decision-making. But that's less than what ET180 is suggesting.