Wyckoff Trading Method: Point-and-Figure and Renko

Point-and-Figure and Renko

In this age of algorithmic and high-frequency trading, point-and-figure charts attract little attention. They occupy a dusty, forlorn place in the library of technical analysis. The earliest works about point-and-figure charting were written by “Hoyle,” an anonymous author, and Joseph Klein. 179 Wyckoff presented point-and-figure charts in Studies in Tape Reading (1910). He used them extensively and most of the chapters in his original (1932) course dealt with the subject. Victor deVilliers, one of Wyckoff’s early associates, published his famous book The Point and Figure Method in 1933. An excellent overview of point-and-figure charting can be found in H. M. Hartley’s Profits in the Stock Market (1981 edition). Wyckoff’s method of selecting stocks began by determining which stocks and groups were in the strongest position in relation to the broader trend of the market. He surveyed the point-and-figure charts of these stocks and groups to determine where the largest amount of preparation existed. In one section of his course, he wrote, “When I was doing my best work, I discarded everything but a vertical line chart of the daily average of 50 stocks, with volume, and the figure charts of about 150 leading stocks.” He added, “Figure charts are more valuable than Vertical [bar] charts.”

In this chapter, I discuss two of the most important aspects of point- and-figure charting:

• How to select point-and-figure box size and reversal.
• How to locate a line of congestion and make projections. 

We have already discussed the construction of point-and-figure charts based on a 1:1 and 1:3 ratio. Figure 9.3 of December 1993 bonds typifies the 1:1 or 1 -point type of chart. Figure 9.12 illustrates the 1:3 or 3-point reversal. In Figure 9.2, I demonstrated a less common 1:2 ratio chart. Wyckoff’s workhorse was the 1-point chart based on dollars per share. Of course, he surveyed every price change to plot these charts. Given today’s volatility, most point-and-figures are made from closing prices of various time periods. When I want to make a point-and-figure chart, first look for areas of price tightness. My study of currency futures led me to the daily continuation chart of the British pound, where prices narrowed into a tight range between August 2012 and September 2011. A quick check of the monthly chart showed this tightness extends over to 2009. Rather than trade futures, a long position in FXB, the ETF of the British pound, seemed low risk for a tax-deferred account and would not involve rolling positions from one contract to another. The daily price rise from the 154.52 low on August 10 indicated a long position was warranted.

The next step involved deciding upon the box size and reversal unit. Let’s try a 1 X 1 calculated from daily closes (Figure 11.1). We immediately find seven very tight columns along the 154 line projecting a rise to 161. (Note that a count for an up-move is always made from a price low; use a price high to project downward.) I like to make the count from the point where the up-move begins in earnest. Some traders I have known would immediately measure across the entire span and expect a rise to the maximum count. I prefer to break the count into phases starting with the most conservative one. One of the easiest ways to do this is to count over to a “wall” where prices accelerated upward or downward. Here, we have four phases projecting a minimum 7-point rise to 161 and a maximum 21-point rise to 175. The average of the four targets calls for a move to 168.75. A move of this magnitude would extend into the band between the 2009—2011 highs. But we still do not know where the up-move will peak. In the end, the point-and-figure projections are guidelines; however, they have an uncanny accuracy. The 11 columns counted across the 164 line called for a decline to 153, which the stock reached one-month later. Notice that prices did not trade in 4 out of the columns combined to make this count. After the stock fell more than $7 from the 164 congestion line, it would have been clear a larger count should be made.

 

Figure 11.1 FXB 1 × 1 Point‐and‐Figure Chart

Next, we see a 1:3 ratio chart made with a 4-point box size and a 12-point reversal (i.e., 4 x 3); it is calculated from daily cash Standard & Poor’s (S&P) prices (Figure 11.2). The congestion across the 1344 line spanned five months between February and July 2011. Its projection for a decline to 1100 exceeded the actual low by only eight points. As of this writing, all but two counts on this chart have been unfulfilled. The largest count consists of 17 columns across the 1160 line in the period between November and August 2011. When the projected points are added to the 1160 line, the target is 1484; however, another count can be made by adding the points to the low point of the count zone. This generated a less extreme 1424 target. Point-and-figure charts are noted for filtering price data and thus showing the broader framework of price movement. It is accomplished by adjusting the price reversal and data field to the point of maximum clarity like focusing a microscope. With practice, one learns how to find the right balance.

Choosing the best box size, reversal and data field for a point-and-figure requires practice. Let’s take the example of Hecla Mining, a relatively low- priced stock, after the close on Friday, August 3, 2012. The weekly chart (Figure 11.3) has a tight pattern between August and January 2012. Weakness in April—May pushed prices below a support line before the upward reversal in the week ending May 25. This can be viewed as a spring. It is followed by

 

Figure 11.2 Cash S&P 500 4 × 3 Point‐and‐Figure Chart

 

Figure 11.3 Hecla Mining Weekly Bar Chart

10 weeks of lateral movement within an 81-cent range as the stock stands on the springboard awaiting a catalyst. To make a point-and-figure chart, I prefer to begin with daily data because they tend to create tighter patterns. A 1: 1 ratio chart will usually offer more lateral movement. A 25-cent box and reversal size may work but these values are about 6 percent of the stock price. A smaller percent will show more price work but only a few keystrokes are needed to change the parameters. Not unexpectedly, Figure 11.4 is unsatisfactory. First, the congestion across the 4.25 line only covers nine columns. The calculation (9 X 0.25) + 4.25 projects a rise to 6.50, a respectable return; but it is not in proportion to the amount of time spent moving laterally. Secondly, the January 2012 low does not appear because the chart is constructed from daily closes and thus filters out intraday lows and highs. Thus any point-and- figure chart made from daily closes will have the same problem.

 

Figure 11.4 Hecla Mining 0.25 × 1 Point‐and‐Figure Chart

Figure 11.5 takes a different tact. It uses a smaller box size and reversal (0.05 X 3) calculated from hourly closes. Anyone familiar with point-and-figure charts would like this setup. Here, we see three separate phases that generate targets of 7.15, 8.50, and 11.20, respectively. 

Figure 11.5 Hecla Mining 0.05 × 3 Point‐and‐Figure Chart

Wyckoff would look at this chart and explain how the composite operator accumulated stock during the eight-month period. Composite operator was Wyckoff’s generic term for the insiders and pools, who profited by accumulating or distributing stock in preparation for a campaign trade. Here, we see that the large operators forced prices below the trading range in April to find out how much supply could be drawn out. The 50-cent upswing at point 1 marked the largest gain since the February high and reflected demand. The next pullback failed to retrace 50 percent of the up-move, a bullish condition. On the rise to point 3, the upward thrust shortened as bids were pulled. The price action between points 4 and 6 shows the composite operator tried to keep a lid on the stock in order to complete his line. From the low at point 6, the stock was in strong hands as the volatility ceased and price rose steadily. I don’t doubt such large forces are at work in the marketplace, but their activities are not the focus of my attention. The count made across the 4.45 line in Figure 11.5 is subdivided into three phases. Count AB covers the price movement from early August to late June. AC stretches leftward to the breakdown on April 10, and AD incorporates all the price work to the January 11, 2012, low. The chart is posted through September 25, 2012, and shows the strong liftoff out of the trading range. Helca peaked on the last up-move shown here at 6.94 just 20 cents below the count AB target. 

If the stock holds above 4.45 on the next pullback, the larger counts may be fulfilled in the future.

Figure 11.6 of the March 2011 five-year note is one of my favorite examples. The minimum tick in this contract is one-quarter 32nd valued at $7.8125. The point-and-figure chart uses a one-quarter 32nd (0.0078125) box size and a reversal 2X greater (0.015625)—in other words, a 1:2 ratio. It is constructed from 3-minute closing prices. Here, the duration of each column (or wave) is plotted on the chart. You see how the early morning low at 11715.5 was penetrated on the sell-off to the final low. The first down-move covered eight ticks in 33 minutes; the second spanned six ticks in 18 minutes; the third equaled three ticks in only 9 minutes. You can see the ranges narrowing and the time lessening exactly as occurred on the wave charts. It reflects no ease of downward movement and diminishing time on the down-moves. The selling pressure is clearly spent. From this low, the market rallies 12 ticks over the next 33 minutes.

Figure 11.6 March 2011 Five-Year Note .25/32 X2 Point and Figure Chart 

On the way to the day’s high, most of the down-waves last between 3 and 9 minutes, with the exception of two spanning 15 and 18 minutes. Both of these down-waves equal the minimal two-tick reversal, which tells another story. Prices rally vigorously over 24 minutes to the top (11724.75). Because they exceed the upper limit of this handmade chart the full swing is clipped short. This is the most amount of time on any up-wave since the contract rose from the low. It assuredly had climactic volume. The next down-move covers a relatively small amount of ground, but the 36 minutes stand out as the largest downtime. Imagine all this time spent without any ability to recover. I think Wyckoff would say the composite operator is trying to support the market in order to establish more shorts. It is an overtly bearish change in behavior and leads to the largest down-move of the day session. Notice that it lasted only 6 minutes in response to a bearish Treasury auction. The last upswing (11722.75) in the top formation lasted 9 minutes before prices plummeted over the next 36 minutes. The 19 boxes across this line project a decline to 11713.25. Prices fulfilled the conservative target of 11715.25, calculated from the day’s high. As an aside, the five-year note is an excellent trading vehicle for low-capitalized and/or less experienced traders. Given the low margin rate and big volume, large traders can easily trade size to make the smaller swings more worthwhile.

It should be obvious that one can substitute time for volume. To make this information more accessible on all point-and-figure charts, a friend created a simple indicator that plotted the duration of each column as a histogram below the price work. Figure 11.7 shows this indicator on a December 2012 silver point-and-figure chart (0.01 X 3) made from one- minute closes. Supply first appears at point 1, but the next up-wave (point 2) tests the earlier high. At this point, the buyers have an opportunity to gain the upper hand. The lack of upward follow and the ease of downward movement at point 3 say the sellers are stronger. Silver then treads water over the next 50 minutes until buying emerges at point 4 where prices hold firm without a 3-cent reversal for 25 minutes. The bullish argument says the buyers are absorbing the selling. Price must continue higher. Instead, silver hesitates for 17 minutes and drifts to point 5, where the uptime is only four minutes. Because of the market’s inability to rise after the action at point 4, we know it encountered supply. The down-wave at point 6 over the next 25 minutes erases the bullish story. The sellers maintain their pressure on the market at point 7 for an additional 36 minutes. Following the break below the low of point 4, an outpouring of selling takes prices lower in very little time. When measured together, all of the congestion along the 34.51 congestion line projected a decline to 33.85. Before the closing bell, December silver reached 33.92.

Figure 11.7 December 2012 Silver 0.01 X 3 Point-and-Figure Chart

Between the lows at points 3 and 7, silver traded for 3 hours and 21 minutes. The action at point 7 is particularly telling as the sellers keep the pressure on for 36 minutes. Twenty-two waves formed during this period, which is much more manageable than the corresponding 201 one-minute price bars. The capability to filter price movement is one of the benefits derived from using a point-and-figure. But I know of no way to determine the amount of time and volume for each “x” or “o” unless a chart is manually maintained, as shown in Figure 9.3. That chart did not provide the minutes for each plot. Renko charts offer this capability making it the consummate tape reading medium. I doubt Wyckoff ever saw a renko chart, but, if he had, its benefits would have attracted his utmost attention. If you look through the books and articles on renko, the same information is repeatedly stated: the Japanese invented renko about a century ago, it is composed of bricks or renga, it shows support and resistance levels extremely well, and it deals only with price without regard for time and volume. Fortunately, computerized renko charts do provide the volume and duration of each brick. Because of this, wave volume can be plotted below the swings on the renko charts. They come closest to re-creating Wyckoff’s original tape reading charts—except he did not show the time between the waves.

 

Figure 11.8 Renko Brick Formation Diagram

Renko charts, like point-and-figure, filter out much of the noise and ambiguities that accompany bar charts. The formation of a renko brick is depicted in Figure 11.8. Assume we are looking at a number of rising bricks with a $1 size. The last completed up-brick in the progression stopped at 10. To form another up-brick the stock will have to trade at $11; to reverse direction and complete a down-brick the stock will have to fall one dollar below the last low to $8. Before the new brick forms, prices can therefore travel within a $2.50 range. The new brick might last 50 minutes before completion. During this time, a five-minute bar chart can give mixed messages that may prompt a trader to close out a trade prematurely or totally miss the coming move. For this reason, renko charts offer peace of mind. They reduce the number of decisions. The buildup of time per brick stems from brick size and the speed of the trading. Rapid price movement causes bricks to last only seconds. Other times, a lengthy brick time may occur as prices absorb through support or resistance levels. Think of a one-point brick in the S&P between 1190 and 1191. Let’s assume it’s the most recent brick in an upward progression. During the next brick’s formation, prices can fluctuate between 1191.75 and 1189.25—2.50 points—for as long as it takes until an 1192 or 1189 print occurs. Naturally, if the brick size is 0.50-point, the time per brick will be smaller and many more will appear; a 3-point brick will obviously span much more time. A day trader in forex or currency futures might use a 5-pip brick (-.0005), while a swing trader could use a 20-pip brick (0.002). One of the distinguishing features of renko is its freedom from time periods. As soon as price fills one brick, another begins. This makes it more akin to Wyckoff’s tape reading chart, where price changes are not tied to set time periods. This also holds true for waves on renko charts.

Figure 11.9 presents a 5-pip renko chart of December 2011 Australian dollar on December 9. Here we see a double top that spanned 50 minutes. The 855-contract volume in the brick at the second peak was the largest to date. It lasted only 7 minutes. On the next up-brick, volume expands to 1,270 contracts over 21 minutes. It spent all that time without making further gains (effort versus reward). When price turns down in the next brick, we know the sellers overcame the buyers’ attempts to take prices higher. Another way of looking at it is to consider that the sellers were selling at the ask price. In other words, they were not selling at the bid, but rather sold to the buyers who were paying up to own more contracts. This is the same degree of distribution Wyckoff observed on his tape reading charts. The buying effort failed to take prices higher, and it is followed by an even larger amount of selling (1979 contracts) on the next down-brick. Here a 22-minute struggle takes place. Given the lack of demand after the last up-brick, the sellers appear in the stronger position. If another down-brick unfolds, the odds greatly favor lower prices and a short position would be warranted. Buy stops should be placed just above the high of the last up-brick. In about 90 minutes the contract reached 1.0103.

Now look at Figure 11.10. I’m almost certain the stock is TVIX in early September 2011. The brick size looks like 20 cents. From the pre-session bottom, all of the lows held at a higher level. Notice the heavy down-volume and down-time in bricks 1, 3, and 6. What is the message? It is exactly the same message as we saw in Figure 11.6 where the five-year note spent

Figure 11.9 December 2011 Australian Dollar Renko Chart  

15 and 18 minutes on two small two-tick pullbacks. Someone was buying. In TVIX, these three bricks underscore accumulation. Wyckoff wrote about this kind of accumulation rather than some static, preconceived model. Just imagine: at point 3 the stock spent 20 minutes as volume swelled to 200k shares and then price rallied 60 cents! At point 6, the volume was over 250k in 90 minutes, and the stock again refused to move lower. The low-volume pullback at point 7 put the stock on the springboard. There is one more dimension to this chart. From the low at point 3, nine waves can be counted to the leftmost downswing. Multiply 9 by 0.20 and add to the low (39.80), for a target of 41.60. Thus, renko charts can be used like point-and-figure charts to make price projections.

My first handmade experiments with renko charts involved plotting the swings vertically so they better resembled a point-and-figure chart. This made the lines of congestion stand out better. It wasn’t long, however, before one of my students devised a renko format where the volume and time are plotted within the brick. Figure 11.11 of December 2012 S&P on September 25,2012,

 

Figure 11.10 TVIX Renko Chart

consists of a 0.75-point brick size. The upper number within the brick is the volume; time is the lower number. Wave volume and duration are entered at the wave turning points. The result is a powerful tape reading chart (compare to Figures 9.1 and 9.3) with time per plot added. It goes beyond anything Wyckoff ever constructed. The top brick took 15 minutes to form, and volume increased to 35k. This large effort failed to produce further gains, thus raising the suspicion that supply had overcome demand. The next two bricks lasted 43 minutes and the volume totaled 86k; it was then obvious that the S&P met supply on the last up-wave. The first large down-wave out of this top drew out 189k contracts over the next 78 minutes, which marked the beginning of a much larger break.

I have included a 0.25 X 3 point-and-figure (Figure 11.12), which shows the entire top on September 25, 2012. The count across the 1455.25 line projected a decline to 1436.50, 1.5 points above the closing low. This chart shows the accuracy of point-and-figures constructed from small, intraday price movement. The only drawback to such a chart is the way it handles the overnight data. Because the price movement is slower at night, the minutes per column can become unusually large, and they tend to dwarf day-session data. Therefore, I adjust the scale, which in essence clips off the larger 

 

Figure 11.11 December 2012 S&P Renko Chart

readings. The result is an indicator with elegant simplicity. Here, the downturn after the upthrust spans 35 minutes, the largest amount of downtime (read volume) since the opening of the day session. The next up-move lasted four minutes. When the S&P fell below the low of the previous down-wave (1454.50), the message was clear: Go Short. In Studies in Tape Reading, Wyckoff wrote: “[The tape reader] must be able to say: the facts are these; the resulting indications are these; therefore I will do thus and so.”  I call it the moment of recognition when you sense a move is about to happen. The realization sweeps you into taking action.

The reading of the S&P 0.75-point wave chart on September 25, 2012 (Figure 11.13), gave the same insight as the message on the renko and point- and-figure charts. Look at the shortening of the upward thrust and the large effort on the last up-wave (59k). The wave volume was the heaviest on the chart up to that point, and price exceeded the previous high by only one-half point. But it’s the large wave volume that drives home the message as the force of the buying encountered a larger force of supply. The sell-off below the previous low (1455), said the die was cast and immediate action warranted. In this instance, the bearish change in behavior (i.e., upthrust and large effort with no reward) was not followed by a low-volume pullback. The market fell three points on 187k volume before having a minor correction to 1452.50.

 

Figure 11.12 December 2012 S&P Point‐and‐Figure Chart

Figure 11.13 December 2012 S&P Wave Chart

For the record, Figure 11.14 shows the December 2012 S&P five- minute bar chart for the same day. I was weaned on hourly and five- minute bar charts and certainly can read the bearish story here. The upthrust/shortening of the thrust stand out clearly. On the last price bar to the high, the position of the close indicates that the market met selling. But there is no increase in volume to tell us the sellers have gained the upper hand. As I have said before, “The true force of the buying is lost in time.” This is not the case in all situations, and there will be times when the five-minute chart provides a better picture of events; however, it occurs too rarely.

 

Figure 11.14 December 2012 S&P Five-Minute Chart

Wyckoff maintained a wave chart of market leaders calculated from their intraday price swings. Originally, it was constructed in a more precise manner but in modern times it has been calculated from closing one-minute or five-minute time periods. Wyckoff showed how the wave chart of market leaders can be plotted alongside the tape reading chart so the waves on both can be compared. The wave chart of S&P futures serves as my indicator for the market at large. I know of traders who monitor the waves in the SPY for clues about market direction. On September 25,2012, there must have been hundreds of stocks with wave or renko charts that flashed the same bearish message as the S&P. I randomly selected a 10-cent renko chart of Union Pacific (Figure 11.15) for that date and the bearish evidence stood out at a glance. Hopefully, you see it. At the turning points on the chart, I have plotted the wave volume (in thousands) and the number of minutes. If we had to choose one brick that told us what to expect, we would have to select the 128k down-brick at 11:54 a.m. EDT. The volume in this one 10-cent brick exceeded the volume on the 60-cent up-wave, where a total of 104k shares traded. The total wave volume on the decline to the 11:54 a.m. low exceeded the volumes on the previous two up-waves combined. So this is the spot on the chart where we know what will happen. The down-move to the 11:54 a.m. low occurred about eight minutes after the decline below 1455 on the S&P wave chart. Yet UNP held for another 21 minutes before it followed the S&P lower. This lag time would have benefited anyone trading UNP.

Figure 11.15 Union Pacific Renko Chart

Larger brick sizes work very well for trading the intermediate swings. For stocks trading above $20 per share, I like to use 30-cent bricks. Figure 11.16 shows 4-point bricks calculated from S&P continuation data. By filtering so much of the intraday noise, this chart makes it easier to hold trades for 20 points or more per contract. The first large up-wave spans most of three sessions on 4.74 million contract volume. One-fourth of that volume emerged in the second brick (1) of the wave and again in the final brick (2). The one served as the prime mover, and the other signaled stopping action. After this climactic action, profits should have been taken as soon as a down- brick formed. Twenty-eight points were captured between these two bricks. The wave volume on the ensuing down-move was less than recorded in the second brick of the up-wave. At (3), the S&P tried to absorb through the over¬ resistance. Only one brick printed in the next down-wave (4), and it had large volume. As soon as a brick formed above the horizontal line, we would have known the high volume indicated the absorption was complete and the buyers were in control. It was the ideal spot to re-establish long positions. The S&P then rallied another 24 points before supply entered the picture.

 

Figure 11.16 S&P Continuation Renko Chart

I mentioned earlier that my initial experiments with renko involved making a chart where the bricks unfolded vertically. The vertical format allowed more price data to appear on the chart as opposed to the traditional diagonal movement. In addition, a line of congestion on a renko chart can be used to make projections of how far prices may travel.

Figure 11.17 is a handmade chart of the March 2012 S&P on December 16, 2011. The brick size is one point. This first effort was literally scribbled 

onto a piece of notebook paper. The volume numbers in thousands of contracts are written at each price. Initially, I did not show the minutes per brick, but these were added later. On the down-move from the high, notice the two bricks where the volumes rose to 182k and 102k, respectively. Here, we had a combined volume of 286k over a 63-minute period as the sellers clearly gained the upper hand. Just prior to the up-move to the day’s high, notice the low-volume pullback (9k) that reflected a total lack of selling pressure and offered an excellent buying opportunity. The point-and-figure congestion across the 1218.75 line, where the 9k volume appeared, projected a rise to 1223.75, one point below the high. Look at the top price, where the brick volume soared to 79k, the largest reading 

 

Figure 11.17 March 2012 S&P Chart (Scanned)

on the chart. I think you can see the usefulness of such a chart. Right now, it’s a work in progress.