Predictive Indicators
By Darrell Jobman ISSUE 704 | APRIL 2008
Technical analysis begins with a price chart, and you have to read the tracks left by the prices on the chart. The problem with charts is that they are subjective. Different people see different things, and you can't program subjective analysis into a computer. Most technical indicators aren't perfect, as they consider only one thing—price action that happened in the past. Most indicators can lag market action, and you as a trader are relying on information that is old. Predictive indicators are used to help solve this issue. They can give you the ability to make some possible price projections, and help you time your trades. I'll give an overview of some popular predictive indicators, including volume and open interest, pivot points, predicted highs and lows, Elliott Wave theory, Fibonacci, and cycles.
Predictive indicators, which could also be called target indicators, tend to be more proactive than traditional indicators. They are still based on past price action, but are more proactive about analyzing the future. If you are interested in more details about a particular type of analysis, I encourage you to do some in-depth investigating.
Volume
Volume is more of a secondary indicator than a primary one, as it tries to assess what price moves during a given period might mean. Volume is simply the number of instruments that have changed hands during a given period. One buy matched by one sell gives you a volume of one contract (even though it involves two trades by two people). Volume is a source of information for all markets but, unlike stocks that have a limited number of shares available to trade, futures volume is virtually unlimited as long as there are sellers and buyers.
In the soybean futures continuous chart below, volume is reflected in the blue bars at the bottom. We can see big volume days in the first three instances that were accompanied by bearish-looking candles (the vertical red lines), then in the fourth case, a bullish white candle.
Look at big volume days, and note the pattern. When a market is trending, it takes the trading crowd a while to catch on at times. Volume rises as they jump on board. If the number of interested buyers exceeds the number of sellers, price will rise until it becomes attractive enough to entice sellers into the market to fill the buyers' requests. As a result, volume increases. However, if there are no willing sellers, then prices may their reach their daily limit and can lock up, limiting trading volume for that day.
In some cases, the market may be very bullish or bearish, and prices move limit up or down early in the day. Volume can't reflect this strong sentiment when trading limits are in effect. You can see that in the soybean chart on the right side, where big price moves are accompanied by low volume.
Open interest is volume's companion. It represents the number of contracts not yet offset by closing a position. Open interest does not exist in stocks but can be an important factor in futures and options and indicative of market sentiment. Closing out positions as first notice day and as contract expirations near can affect normal price action.
Soybeans Continuous
Source: VantagePoint Intermarket Analysis Software
Analyzing Volume and Open Interest
The real significance of volume and open interest is in the relationship to price and price action. If volume and open interest are high and rising, the trend is strong. If they are low and declining, the trend is weak. The table below shows the possible volume and open interest combinations and effects they might have on market value. Note that the two bearish combinations show prices can be up, but if volume and open interest are down, that indicates declining interest among market participants. If volume and open interest are up and prices are down, that means more traders are jumping in to establish short positions. Eventually a seller will find a buyer, but price could be lower. Also note the last bullish line. Even though prices are down, a decline in volume and open interest suggests fewer positions as prices move lower, and that's bullish.
| Price | Volume | Open Interest | |
| Bullish | Up | Up | Up |
| Neutral | Up | Down | Up |
| Bullish | Down | Down | Down |
| Neutral | Down | Up | Down |
| Bearish | Up | Down | Down |
| Neutral | Up | Up | Down |
| Bearish | Down | Up | Up |
| Neutral | Down | Down | Up |
Volume and open interest do have problems as a method of price prediction. Volume figures may not be available until the next day, too late to use for overnight analysis. Also, typically traders look at total volume, but sometimes price action for one contract in a market may be significantly different from another contract in the same market. Unless you look at volume for individual contracts, volume analysis may not help you determine how volume impacts price of the contract you are trading. As previously mentioned, volume may also be restricted by price limits, and not reflective of market sentiment. Volume may also become distorted as hedge funds and other large traders flood into commodity markets. It could make volume grow, but not be indicative of what price fundamentals ordinarily should reflect.
Who is producing the volume? You can find a breakdown of where various participants in the market (small speculators, hedgers, funds, etc.) are positioned and to what degree in Commodity Futures Trading Commission's Commitments of Traders report, released weekly. (Visit www.cftc.gov)
Pivot Points
Volume confirms general price direction, but we want to look at indicators that attempt to predict specific price targets. One of these is pivot points. Why pivot points work is subject to debate, but they've been around a long time and many traders respect them, so they can become self-fulfilling prophesies. The first calculation of the pivot itself is found by adding the high, low and closing prices and dividing by three.
Pivot point (P) = (High+Low+Close)/3
Now you can calculate support and resistance based on the pivot, using the formulas below. There is some flexibility in determining your pivots, but generally, I recommend going with the regular day session, because that's what most traders use. You can also use different time frames, like weekly or monthly price points. The pivot acts as the average price for the period, and the resistance and support points act as your trading range.
Resistance 1 (R1) = (P X 2) – L
Resistance 2 (R2) = P+H – L
Resistance 3 (R3) = R1 + (H-L)
Support 1 (S1) = (P X 2) – H
Support 2 (S2) = P – H+L
Support 3 (S3) = S1 – (H – L)
Professional traders usually trade against the boundaries in the range, but breakouts of these boundaries can also act as clues for possible change in trend.
The chart of the E-mini S&P 500 futures below provides an example of pivot points in action. The light black line represents the pivot (the high, low and close of the prior day divided by three). The R1 and R2 lines above and the S1 and S2 lines below represent boundaries of expected trade. The first arrow on the far left shows where the S2 line turned back prices. The next set of arrows shows how the primary R1 and S1 (dashed blue lines) contained price action during that day. The next arrow showed how prices dropped though S1, then once that happened, rebounded from S2.
The basic approach to trading based on these levels is to have a limit sell order below the R1 line, and a limit buy order above the S1 line. You are trying to fade a move to those lines because that's where the market is likely to turn. In case the R1 and S1 might not hold, you would place a buy-stop a short distance above the R1, and sell-stop a short distance below the S1 line. Then if prices do break through your first support or resistance, you'll be stopped out with (hopefully) only a small loss. You may want to then repeat that strategy at the R2 and S2 lines. This technique does depend on keeping an eye on the market throughout the day as you trade.
E-mini S&P 500 with Pivots
Source: eSignal
Intermarket Analysis Predicted Highs/Lows
One approach that's similar to pivot points involves intermarket analysis, using data points calculated in VantagePoint Intermarket Analysis software. The data reflects the relationship or influence other markets have on a target market. VantagePoint uses the same data points as pivot point analysis: the sum of the daily high, low, and close divided by three. This produces a typical price, and actual or predicted prices help to forecast the next day's predicted high and lows. These predicted highs and lows form a band or envelope within which the market is expected to trade. As with pivot points, the first requirement for an intraday trade is that the opening price is within the boundaries of the range, preferably in the middle. The trader could try to fade the predicted highs and lows by placing limit orders several points inside the boundaries. Those points are indicated by the green horizontal lines on the chart below of the E-mini S&P 500 futures.
Stops are placed just outside the boundaries (the purple lines). Whether you get long or short, or whether you even have a trade, depends on whether or not prices open inside the range, and the time of day when the market reaches your limit buy or sell order.
On the first few days in this E-mini example, prices opened outside the calculated parameters, or you got stopped out with small losses. I prefer to use prices during the regular session for my analysis. That means some days you may miss trades if a move to a predicted high or low occurs in overnight hours in contracts that trade around the clock, like the E-mini S&Ps.
In this example, I used two points above the low and two points below the high for the limit orders, and two points above the high and below the low for the stop orders. If a trade were entered exactly at the limit and exited at the stop, the risk would be about $200. Keep in mind, a market can have a much wider range. We aren't attempting to capture the whole range on a trade, just a piece of it.
Bar two, the black bar on the chart below, is no trade because prices didn't approach either limit order until the close. Bar three would not have been a particularly good trade because it would result in being stopped out at the high, before the market moved to your lower target.
Using the guidelines mentioned, this strategy did work on the first bar, when we would've gotten long, and bars four or five, when we would've gotten short. The net gain on all five days would be 25-26 points, assuming you got all fills at the prices you specified.
In volatile periods, there may be many days using this strategy where you won't have any trade at all because prices will be outside your parameters.
E-mini S&P 500 Intermarket Analysis
Source: VantagePoint Intermarket Analysis Software
Elliott Wave
We've looked at short-term indicators for intraday trading, now let's look at predicative indicators that can be used for longer-term trading, even years out into the future.
Elliott Wave was developed more than 60 years ago by an accountant named R.N. Elliott, and his theory is a course of study in itself. The basic premise is that markets move in a series of waves, not necessarily due to supply and demand, but due primarily to the collective mass psychology of traders who will respond the same way in similar circumstances and produce a rhythm that repeats itself over and over. It's more a study of human behavior than prices, and the key is to read the mood of the crowd.
A complete pattern has eight waves. An uptrend will have five waves up and three down. Waves one, three and five are impulse waves that rise with the trend. Wave three tends to be the longest and strongest. Waves two and four are corrective waves within the rising pattern. Following the five-wave pattern is usually a three-wave ABC corrective pattern. There are waves within waves within waves, major and minor waves, and labeling them can become a real challenge. As conditions change, you have to adjust the wave sequences. That's not even getting into zigzags and other patterns that are also part of Elliott Wave analysis. There is more to the theory than we can discuss here, but you can get a general idea of the concepts and see if it's something that piques your interest.
On the next chart of the Nasdaq Composite, the large numbers show a wave pattern that's of a higher degree than the smaller numbers. The four on the left and five on the top indicate the completion of a previous larger wave, and the one on the lower right indicates the beginning of a new pattern of the same degree.
Within the larger degree pattern, there are several sets of smaller numbers identifying the waves of smaller five-wave completed patterns. Within wave two on the right side is yet another smaller wave pattern, numbered with Roman numerals.
Nasdaq Composite
Source: VantagePoint Intermarket Analysis Software
In theory, Elliott wave seems to lay out the future for you. However, in practice, it's difficult to first comprehend all the nuances of the theory, master it, and apply it on a real-time basis to trading.
Fibonacci
The basis of the waves in Elliott Wave theory goes back to an Italian mathematician in the 13th Century, named Fibonacci. Let's explore Fibonacci numbers and ratios and see how these two approaches work together.
Starting with 1+1 = 2 and then adding that sum to the previous number, Fibonacci developed a sequence of numbers and ratios between numbers that Elliott dubbed “nature's law.” Once you get into the sequence a ways, you will discover the ratio of a number to its next higher number is always around 0.618. Conversely, the relationship of a number to its next lower number is always about 1.618. The number sequence and ratios are below.
1+1 = 2, 2+1 = 3, 3+2 = 5, 5+ 3 = 8, 8+5 = 13, 13+9=21, 21+13 = 34 and on and on...
Major Fibonacci ratios 0.382, 0.50, 0.618, 1.0, 1.618, 2.618
The relationship between these numbers and these ratios has been observed in everything from the pyramids in Egypt to the coils of seashells. All of these numbers and ratios may seem mysterious, but it's an interesting way to study the markets.
Elliott's wave patterns involve Fibonacci numbers like 8, 5, etc. The length of many waves relative to other waves boils down to Fibonacci ratios. In an ideal Elliott wave pattern, waves one and five would be about equal in time and price. Wave three would extend at a ratio of 1.618 the magnitude of waves one and three. This analysis can help a trader predict duration and length of a price move.
Fibonacci Retracements
Going back to the Nasdaq chart, the move down from the peak at 2861 to the low of 2540 totals 321 points. Assume you were watching the bounce back and wondering how far it could go. Once the market moves above the 0.38 Fibonacci level, a 50 percent retracement becomes your target. Then once it moves above that, the 0.618 retracement becomes your target. That Fibonacci line did stop the corrective advance, and the market did turn lower after hitting the 0.618.
Nasdaq Composite
Source: VantagePoint Intermarket Analysis Software
When the market has endured a setback, participants want to know how high a corrective rally might go. The weekly E-mini S&P chart below shows some projected targets using Fibonacci retracements. The market had made a 25 percent retracement of the distance from the October high to the March low, making the 0.38 level at 1381 the next target on this chart. That's in the same zone as the resistance from the lows in 2007 (red dashed line). That area could become a tough obstacle to get through, but if it does, then a 50 percent retracement becomes the next target, near 1420. This isn't my prediction for this market, but is meant to show how Fibonacci might be used to find price targets.
E-Mini S&P Weekly
Source: eSignal
Fibonacci Extensions
Another use of Fibonacci numbers and ratios is in extensions, which predict potential price targets for upcoming waves. Looking at the Nasdaq chart again below, let's assume we are coming off a correction high (the top red dashed line, the one hit when the market rebounded to the 0.618 line) and prices are moving lower. The question is, how low can prices go, especially when the low at 2540 breaks? Recall the difference from the top to the first low was 321 points (numbers in blue). Using a Fibonacci ratio of 1.618, we come up with 519 points. Subtracting that from the correction high of 2735, we arrive at a target of 2216. The market did reach that target level quickly, and then dropped slightly below it. That target price was a nice piece of information to have.
Nasdaq Composite
Source: VantagePoint Intermarket Analysis Software
Pairing Elliott Wave and Fibonacci Analysis
Let's pull together the Elliott Wave and Fibonacci concepts. In this next chart, the weekly July 2007 soybean futures, we can see the huge rally from the low around $6.30 a bushel in September 2006 to the high of $15.95 in March 2008. The numbers with the question marks on the chart are just my observations; my waves might not meet all the necessary criteria to be the true Elliott Wave count. (I have seen my five labeled as a three.) Wave one represents a price increase of slightly more than $2, and wave three just under $2, so we'll use $2 as a starting point for projecting the length of wave five. As the rally got rolling in fall 2007, we may have concluded this could be a wave five. If waves one and three are about the same, wave five should extend. Using 1.618 x $2, that's $3.24, and adding that to the May 4 low gives us a target of $11.70. Prices quickly broke through that level without a stir.
Let's project a little more. Take 2.618 x $2 and add that to the low, and our target becomes $12.70. Again the market zipped through that easily. Using the next level, 4.24 x $2 we get $8.48, and the target becomes $16.90. The market hasn't made it there–at least not yet–but the growing season is just ahead.
Now let's assume that we've seen the top, and notice that big black bearish candle in wave five. If you want to tackle the short side, how low do you think the market could go? A 50 percent retracement of the wave five move would take prices down to $12.20, and that's where the corrective setback stopped initially (see the blue arrow). When the downtrend resumed, the decline stopped at the 0.618 Fibonacci ratio, as the red arrow points out.
Soybean Futures
Source: eSignal
On this next chart, the letters are just for my reference only, not part of any Elliott Wave analysis or indicative of an A-B-C correction. Say the market has come out of a low at point C. You may be wondering whether it's prudent to stay in a long position after such a lengthy and almost parabolic move. You know the difference between points A to B is about $5.30, and if a move from C to E (unknown) were to match that, the target would be about $17.50. If you take 0.618 of the move from A to B, you get $3.27. Adding that to the point C low gives you a target of $15.50. Prices have exceeded that, leaving the next target $17.50. We'll leave that target hanging out there and see what happens.
Soybean Futures
Source: eSignal
Cycles
If you've been baffled by waves and ratios, cycles may be easier to grasp. It's pretty much economics 101, based on the dynamics of supply and demand. As demand increases, price rises until supply can meet it, then prices fall. So high prices beget low prices, and vice versa. The length of time this takes to occur depends on how long it takes the market to absorb changes in supply and demand, and the type of market it is.
When we think of seasonality, we often think of agriculture, but it also applies to other markets. We know the summer driving season usually increases demand for gasoline, and gasoline prices tend to rise in the spring and summer. Even if you don't know much about technical analysis, knowing about seasonal and cyclical market patterns can be useful to your trading. Of course, you still have to determine timing of the cycles. Cyclical highs and lows might come in a broad time frame, and that makes it more difficult to use as a tool for short-term trading.
As an example, on the hog futures chart, the market's lows seem to come every four years or so. The tops also occur in between the lows every four years. The cycles aren't so clear more recently, perhaps due to shifts to commercial hog production.
For cycles of this magnitude, the time period shown is not nearly long enough to determine the length of a cycle. You have to go back many years to determine if the cycle is a reliable pattern. The goal of a cyclical analysis will put most of the focus on the timing of highs and lows. Then you would bring in other techniques to project potential highs or lows.
In the hog chart, you can see prices tend to make seasonal highs in the May – June period (blue arrows). There is logic to this pattern. Farmers are busy planting corn and soybeans in the spring and time the hog production cycle to have the fewest hogs in the spring and the most hogs in late fall or winter, when feed is more available and cheaper.
But be careful how you read this continuous futures chart because there is typically a price discrepancy between contract months. When the typically lower-priced winter hog contracts expire, they are replaced by the typically higher-priced June hog contract. You can get a bump in the continuous chart that may be misleading when contract changeover occurs.
As continuous charts make the transition from contract to contract, they might leave gaps. In some markets, use these types of charts with caution.
Lean Hog Futures
Source: eSignal
Long-Term Cycles
There are not only seasonal cycles, but also long-term, multi-year cycles. The next chart shows a 44-month cycle in cocoa futures. The lows in the red boxes fall right on schedule--until 2008. Instead of an expected cycle low, we saw an unexpected high. The speculative boom in commodities seems to have taken cocoa futures for the ride higher. As we can see, no method is fail-proof!
Cocoa Futures
Source: eSignal
In sum, all indicators have value. Hopefully this article has piqued your interest in learning more. Whatever method you decide to pursue, do your own research and choose what fits your method and style best. Predictive indicators can give you an edge, but keep in mind, they are subjective. And, they are dynamic. As conditions change, signals and patterns may change, and you will have to keep adjusting.
Darrell Jobman is Editor-in-Chief of TraderEducation.com, a Web site that provides free information and education for traders. It includes daily and weekly market commentaries, tutorials and a number of other resources for traders. Darrell Jobman can be reached by email at Darrell@tradingeducation.com.
This article was based on a webinar presented April 9, 2008. You can view an archive of this webinar at no charge, as well as others in this technical analysis series presented by Lind-Waldock and TraderEducation.com. Part six of this series will be presented live on May 14. Just visit Lind-Waldock's Events page to view the archived webinars or sign up for the next event at no cost.
Kristina Zurla Landgraf is editor of Lind eWire. She can be reached by email at editor@lind-waldock.com.
Futures trading involves substantial risk of loss and is not suitable for all investors.
Past performance is not necessarily indicative of future trading results. Trading advice is based on information taken from trade and statistical services and other sources which Lind-Waldock believes are reliable. We do not guarantee that such information is accurate or complete and it should not be relied upon as such. Trading advice reflects our good faith judgment at a specific time and is subject to change without notice. There is no guarantee that the advice we give will result in profitable trades. All trading decisions will be made by the account holder.
© 2008 MF Global Ltd. All Rights Reserved.
