Option Investor
Trader's Corner

Dow Theory + Moving Averages: which to use

Printer friendly version

I am writing the first part of my sort of 'annual' column on Dow Theory and how it relates to the basics of technical analysis, which is a large enough topic to require more than one column. 

It seemed timely to me since I noted in my last Index Trader column ('Always Different, Always the Same') that last week's sharp sell off could be a top or cap to the most recent advance, which was a retracement of part (2/3rds) of the last decline in the Dow; at a minimum, this last downswing was a tradable trend using Index puts with an exit probably by Monday; old trader's saying: "take quick profits" (when they come). 

The Market of course came back some yesterday (TUES), but it remains to be seen where this goes. By the way, ever notice how Monday, Wednesday and Fridays are often continuations of the existing trend and Tuesday and Thursdays are often 'reversal' days? But that's another story. 

Getting back to a second point I made in my last Index Trader column about a possible top: one noticeable feature of the recent advance in the Dow Industrials (INDU) was the fact that the Dow Transportation Average (TRAN) didn't follow suit with a similar strong rebound. 

According to Charles Dow and his so-called 'Dow Theory' (he never called his body of writings a 'theory'), the recent advance in the Dow 30 (Industrials), with the Transports following suit, makes the Dow 30 rally suspect about carrying on to a NEW high above 11,000. If INDU did go to such a new high and TRAN did NOT, this would be a definite 'non-confirmation' and possible sell Dow theory 'sell signal'.

This market juncture seems like a good time to lay out the main tenets of Charles Dow's observations of market behavior and how 'technical analysis' useful in options trading, developed from and after Dow's work. At least in the west; Japan was another story. 

Before going further on Dow, this response is to an e-mailed question on moving averages from an Option Investor (OI) Subscriber: 

I enjoy your Index Trader column but I am confused because different analysts talk about using the 10, 20, 30, 50, 100, and 200 day SMAs (Simple Moving Averages) and/or the 10, 20, 30, 50, 100, and 200 day EMAs (Exponential Moving Averages) That makes 12 different moving averages. 

For picking entry and exit points to "swing trade" index options (QQQQ, SPY etc.), which moving averages do you recommend using?

Well thanks for your note and indication that you find value in my Index Trader column (found ONLINE on the Option Investor home page by clicking on "Index Trader").

This subject of which moving average to use and why, is likely worth an Trader's Corner column of its own and I'll get to that in one of my future Wednesday columns (and perhaps say more about why I use particular moving averages in my Index Trader weekend articles from time to time).

I myself use simple moving averages almost exclusively. I am not just not looking for the quicker 'trigger' that can be supplied by the Exponential Moving Average (EMA) 

For those that don't know, a 'simple' moving average adds together a certain number of time periods (e.g., for 10-days of the closing price of an index, stock, etc.), whether the period (or, 'bar') is days, hours, weeks, etc. Each period in a simple average like this is EQUALLY weighted; e.g., in a 10-day simple moving average, each of the last 10 days' closes is simply 1/10th of the sum total that is divided by 10. Simple. 

The 'exponential' calculation for an average allows recent price activity to generate a more rapid change in an average price. A type of "smoothing" is applied; one that assigns a percent value (for example, .15), to the last close and this value will be added to a percentage of the previous period's close. 

The higher the percentage weighting (e.g., 15, 25 or 50%) given to the most recent close, the more sensitive will be the resulting moving average to the most recent price change(s). All data previously used is always part of the new result, although with diminished significance over time. 

The answer to your question also very much depends on what time frame you are trading: what it means that you are a 'swing trader'? Day to day, trading every, or most minor price swings or moves that develop? If so, an EMA (Exponential Moving Average) may be a quicker 'trigger' so to speak and alert you a change in trend quicker. This is a consideration of the TYPE of moving average. The next topic is 'LENGTH' considerations.

Use the 5 and 10-day moving averages on daily charts if you are a short-term (swing) trader of options to buy calls/puts, or buy/sell short the QQQQ Nas 100 tracking stock. For example, by taking a position if QQQQ crosses above/below its 5, then, 10-day moving average. However, I attempt to figure out or 'time' intermediate trends and trend changes, in order to profit from larger price moves. Therefore, I use moving averages of at least 13 or 21 days. 

Another part of my response is that I use 13 and 21 day 'length' setting because these numbers are part of the 'Fibonacci' number series (1,3,5,8,13,21, etc, where each number is the sum of the prior TWO numbers). Why? 

This would be a long answer, not fully covered here. You can go to a prior Trader's Corner (3/30) on the use of Fibonacci numbers in "Fibonacci retracements" by clicking here

But figure I owe you a longer article on moving averages, including use of the fibonacci numbers for the 'length' setting for moving averages and why that could be advantageous. 

For example, as long as the indices are trading above the 21-day SMA, I assume the up trend will continue. (The fibonacci number '21' is where the 20 comes from actually.) Pullbacks TO or around the 21-day moving average will often tend find support or buying interest in this area. If not, and the 21-day SMA is pierced, I look for the near-term trend to be down in the period ahead (over days); if so, I look for rallies up TO the 21-day SMA to get deflected and for selling to come in as this average acts as 'resistance'; if so, this becomes a put-buying opportunity. 

As hard as it is completely generalize on this subject, I would say to forget about the longer numbers for you as a shorter-term trader EXCEPT perhaps to take note of the 50 and 200-day moving averages. When a closing price first (after some weeks) crosses above or below the 50-day moving average, it's an ALERT that the intermediate trend direction may be reversing. 

Crossing above/below the 200-day moving average (some prefer the 100-day, but the 200-day SMA is in more widely used and watched) often acts as CONFIRMATION that the trend has changed. However, closes above/below the 200-day SMA should then continue. Further, if the 50-day SMA goes on to cross above/below the 200-day SMA, that crossover can be confirmation that the long-term trend has shifted from up to down, or down to up.

If you have some indication that the intermediate to long-term trend has changed, you may or probably will want to trade more heavily in the SAME direction as the longer-term trend; e.g., in a dominant or overall up trend, buy calls more actively on pullbacks to support.


Where the Dow Transportation average (TRAN) has not been following the path of the Dow 30 (INDU), is illustrated by this recent chart, one I used recently in the (Index Trader) column I mentioned and of interest here and topical. This chart will illustrate several things, not only later in terms of Dow Theory, but what came before, on moving averages: 

The upper index in the chart below is the Dow 30 (INDU) of course. The moving average is the 21-day period and is the simple moving average (SMA) discussed already. The moving average envelope study simply applies lines that are 2 to 2.5% above and below, respectively, the day's SMA. The Stochastic also uses a 'length' setting of 21. (It's often the SECOND time the (Slow) Stochastic goes to the upper or lower extreme that marks the best trade in index calls or puts.) These aspects of the chart are not the point of the comparison of INDU and the Transportation average (TRAN), below, but I mention them in passing as relevant to the moving average discussion. 

The most noteworthy point on the INDU versus TRAN charts above is illustrated by the (light blue) trendlines that slope in OPPOSITE directions. The trend in the Dow 30 was up during its last upswing, versus the down trend seen in the Transportation Average (TRAN). Charles Dow, and o talked about his two averages needing to 'confirm' each other; if they don't, the push down or up in either average may not last. This was certainly the case above, at least on a short-term basis and so far in the unfolding of the market trend. 

Strictly speaking, Dow in his articles was speaking about a new HIGH or a new LOW being made, and whether a similar new high/low was achieved in the other average. 

The most recent peak in INDU was about 3 percent under its 12-month high. The recent peak in TRAN was 6 percent under its yearly high. This situation suggests that, assuming the Industrial average does push up to a new peak at some point in the weeks ahead, to watch what happens to the Transportation average. 

We can say definitely that this average (TRAN) would have significantly further to go to catch up the Dow at that point, especially in the face of rising fuel costs. More on the interplay of the two averages and what this business of the interplay of the two relates to later. 


Charles Dow didn't consider his ideas on the market as a "theory", rather his observations on how the market behaved or would behave in terms of the averages he invented. 

What came to be known as "Dow Theory" is not a system of market timing really but more of a forecaster of the major or 'primary' trend (over many months, sometimes years)and a predictor of

Back in the 1880s and 1890s, Charles Dow (who, along with Edward Jones formed Dow Jones & Co.) came up with the first stock market averages, which became, over time, the Dow Jones Industrials (now 30 stocks) and the Transportation average (Symbol: TRAN and now 20 stocks) as well as a Utility stock average (now 15 stocks). 

As I noted before, the Dow Industrials might be better called the "Dow 30" as these stocks have become more technological, communication, manufacturing and service oriented and less "industrial", unlike the case of the heavy industry stocks like U.S. Steel that were part of the early Dow. 

This average of 30 stocks is not capitalization weighted, as is the case of the Standard and Poor 500 or Nasdaq Composite index. Dow stocks of companies that have become price laggards, even if theyre much smaller companies than say General Electric or Microsoft, can have more of a dragging effect in the PRICE weighted Dow 30 average than indexes that give more weight to the biggest companies with far more shares outstanding as in the S&P 500(SPX). 

Only the Dow Industrials and Dow Transportation (then a group of railroad stocks) averages are used in what became known as 'Dow theory'. 

In a second part to this next week I can cover the main tenets of Dows observations on market behavior, which is a lot more than just how these two averages behave. 

One of Dow's most important contributions was the idea that "confirmation" of the primary trend occurs by the actions of BOTH the Industrial and Transportation averages. A related aspect to this, really the flip side of it - is the concept of "divergence". Dow spoke mostly about confirmation divergences between averages and between prices and volume or between price action and indicators is mostly what came in this century by various technical analysts. 

Dow said that if the Industrials moved to a new closing high or low, without the Transportation average following suit at some point (within a few months usually) and fail to "confirm" the new high or low or, if the Transportation Average (TRAN) goes to a new peak or new low, without the same action in the Industrials we should be on alert for a possible change or reversal of the primary trend. 

The reasons for this are simple, but it was at the time a quite unique and original observation on Dows part. Take the example where industrial/manufacturing activity is strong and the Dow 30 average continues to move higher because of it. But, at the same time, orders for those goods are slowing. This situation results in a build up of inventories. 

Where such a slowdown would show up however, is in transportation activity as there is less shipping business for these companies. Slowing orders in the transportation sector will tend to result in a fall off of company earnings. Astute followers of these stocks would notice this and selling would start to show up in these stocks, either keeping a lid on stock prices or actually driving them lower. 

Conversely, manufacturing could start picking up but might not be at first reflected in a pick up in those stocks as earnings tend to lag orders however, an increase in shipping might be noticed more readily and cause transportation stocks to begin rising. During this period the Dow Industrials might fall to a new low, but not the Dow Transportation average. 

Therefore, a new high or low in the Industrial average, not confirmed by the Transportation average is suspect. In the case of a new high not confirmed by the Transportation stocks it may indicate that the same slowing of earnings and hence stock prices, will show up later on in the Dow Industrials. 

In my book (Essential Technical Analysis) I used the following chart example of a major blockbuster type Dow signal that could have put anyone on the right track in terms of getting into stocks big time as a new low in the Transports was not confirmed by a new low in the Industrials - 

We can assume that manufacturing was holding relatively steady perhaps there was not a big build up of inventories - but transportation stocks were suffering more relative to the heavy volume and good earnings that they were experiencing in the prior year(s). 

MORE, next time - 
Good Trading Success! 

Please send any technical and Index-related questions for possible use in my next Trader's Corner article to Contact Support with 'Leigh Stevens' in the Subject line.

Trader's Corner Archives