Introduction: Sell high, Buy low?

Many years ago, I was teaching a group of brokers from Oppenheimer at the New School in New York, and was asked, "Are point and figure charts are always bullish at the top and bearish at the bottom?" I answered "No, most investors are often deceived into thinking that stocks making new highs are always good and stocks making new lows are always bad; but for the disciplined p&f trader there are some good chart signals that are bearish at the highs." In this article, we are going investigate one of three of these bearish top signals: The Buying Cimax.

You can look at the other 2 Signals in the articles The Broadening Top and The Bull Trap in the Point & Figure College section of the Investors Intelligence ’University’.

The Buying Climax

Buying Climaxes come about when a stock makes a new yearly high, but ends with a loss for the week. We annotate this information on the top 1,000 stocks we follow closely and report them each week on Chartcraft.com under the Buy/Sell Climaxes section.

Buying climaxes are often the first sign that a hot stock is running out of steam and therefore provide not only a useful warning signal for a coming change in trend but very often signal the top itself.

Buying Climaxes are a sign of distribution. We note that most of the time (around 80% of the time), sellers at buying climaxes are only seen to have done the right thing four months later. Many stocks make their highs on buying climaxes.

1. Exceptions to the rule
Not all Buying Climaxes provide meaningful signals and it worth noting the characteristics of a current uptrend. We will eliminate the following cases:

• Signals on stocks which have a small yearly range. In this case, buyers near the highs will not be under the same pressure to liquidate their holdings as they would in a share which has already run up say 30% that year creating a group of nervous investors who have been left "holding the bag".
• Stocks which have a low weekly volume at the time of the climax as this would suggest a lack of participation at this potentially crucial price level.
• The first sell signal in an uptrend is often unreliable. If the climax high is above previous peaks, the stock will probably be showing a longer term bullish pattern of higher highs and lows and the climax may therefore be of short term significance before the long term uptrend re-asserts itself.

2. Example of a buying climax
EPIQ Systems (EPIQ) in November 2001.

The stock made a new 12-month high at the end of November, 2001 at $26 and then ended the week with a loss, closing at $22.60.

Did this stock meet our selection criteria?
Yes, the yearly low was $6.66, so it had a substantial price range for the year.

Volume was sufficiently high with1.5 million shares traded that week.

Someone who bought the stock near the highs would already be nearly $4 offside and since the stock was already up almost 4 times from its yearly low would feel exposed and left "holding the bag".

Previous action in this stock showed a high at $24.66 earlier in the year, followed by a high pole top at $22, followed by a collapse to $10 in September, followed by the rise to $26.

With a P/E at around 66 and no dividend, investors may easily decide to take profits. Chartcraft would have been ahead of them!

3. Yearly Highs or All-time Highs?
There are two types of buying climax: one failing at all-time highs and the other failing at yearly highs.

(a) Buying Climaxes at Yearly Highs
Diebold (DBD) is a good example of a stock which has regularly signalled buying climaxes at the year’s highs. In the past, it has given the following signals:

June 1998 - climax high at $24 moving to $19.5 in six months.

June 2000 - climax high at $32 moved to $23 in four months.

Dec 2000 - climax at $34 led to a pullback to below $30 within one month.

DBD validates the point that 80% of the time sellers in buying climaxes were seen to have done the right thing within about four months.

The most recent buying climax in DBD came at the end of November 2001 when the stock made a 12 month high at $41.10 and closed at $38.79 for a weekly lo of $1.06.

This was just about double where it had the selling climax low in December 1999 at $20. (keep in mind that when a stock rises 50% the P/E also rises 50%) This level is well below the all-time high of $55 made in Feb 1998. Fundamentally, we show DBD with a P/E around 21 and a yield of 1.4%.

(b) Buying Climaxes at an all-time high

The second kind of a buying climax comes when the stock makes an all-time high and then closes down for the week.

You should always keep in mind that there are "smart money" investors out there who will always be competing with you. These include smarter groups such as the insiders, some mutual fund families, people who buy in "Green Zones", and Specialists. There are also not so smart groups like short sellers, some momentum players, people who buy in "Red Zones" and, perhaps, some chartists.

At the all-time high you can be certain that all sellers are making a profit. That is a given. This can be a little or a lot, but it is definitely a profit. The buyer, on the other hand, can either be a new purchaser, or someone "averaging up’, or a short seller covering what is certain to be a loss. The short seller covering has also proven himself not to be so smart like the new purchaser has just "discovered" this stock only after it has gone to an all-time high.

A stock that met this criteria in early December 2001 was Ball Corp (BLL) which made a new all-time high at $70.78, but closed the week at $68.52 for a loss of 64 cents. As you will see from the chart below, this signal has not yet proved itself and after another high pole, the stock is now aggressively testing new highs. We watch with interest!

However, Ball’s history of buying climaxes has also been impressive:

July 1995 at $38 - this proved to be the top and the subsequent down move did not end until nearly two years later in February 1997 at $24.

May 1999 - buying climax at $59 followed by a high pole top signal at $55. The down move bottomed in the "Green Zone" area at $26 in March 2000. From that level the stock has almost tripled to the $70 high.

4. Confirmation from the Insiders

Usually, we like to see some insider selling with these types of extreme highs and so far we have not in Ball. We actually had pointed out a switch to the buy side by insiders last July at $49 and they have been right so far.

Keep in mind, that in a buying climax the seller or sellers have enough power to turn around this stock that is making new highs despite the big demand for the stock. We consider both Diebold and Ball as "Sells", but are probably not yet short sales.

5. Using the Weekly Climax Count for Market Timing

At Chartcraft.com, we report the previous week’s climaxes on Monday in tabular format. It is worth looking out for large numbers of climaxes as they very often signal important turning points in the broad market indices.

A large number of weekly buying climaxes is a bad sign. We currently regard about 125 climaxes as a significant number. The maximum number of buying climaxes in 2001 was 150 which was during the week ended May 25th - this turned out to be the yearly high.

Conversely, the maximum number of selling climaxes in 2001 were 137, the week ended September 28th, the low for the year. We also saw 127 selling climaxes the week ended March 30th, one week before the previous low for the year.

6. Most recent market call

On Monday January 14, 2002, we reported 99 buying and no selling climaxes; this was a useful "early warning" signal to us that the markets were peaking in the short term and this signal has subsequently been followed by a market pull-back. Listed below are the climaxes for that week:-

NYSE (52)
Airborne Freight (ABF) $16, Accenture Ltd (CAN) $28, Allied Capital (ALD) $28, Ackerley Communications (AK) $18, AO Smith (AOS) $20, Amrep (AXR) $6 ½, AirNet Systems (ANS) $9, Best Buy (BBY) $77, BRT Realty Trust (BRT) Chittenden (CHZ) $35, Chicos FAS (CHS) $42, Chesapeake Corp (CSK) $30,Conectiv (CIV) $24, CSK Auto (CAO) $10, Domtar (DTC) $10 ½, Dal-Tile (DTL) $23, Ennis Business Forms (EBF) $10, Ethan Allen (ETH) $43, FDX $55, Fairchild Semiconductor (FCS) $30, Gencorp (GY) $14 ½, Harman Intl (HAR) $47, Hanover Capital (HCM) $8, Hudson Utd. (HU) $30, Hubbel B (HUBB) $31, IBM $126, Ivex Packaging (IXX) $20, John Hancock Financial (IHF) $42, Kirby Corp (KEX) $29, LaFarge Corp (LAF) $39, Mobile Telesys (MBT) $40, Moodys Corp (MCO) $41, Magna Intl (MGA) $69, MI Shottenstein Homes (MHO) $51, Minerals Technology (MTX) $48, Mohawk Inds (MHK) $56, One Liberty Pptys (OLP) $15,Pactiv (PTV) $42, Polaris Inds (PII) $59, Pier One Imports (PIR) $19, Regal Beloit (RBC) $24, RPM $15 ½, Roper Industries (ROP) $52, Rex Stores (RSC) $29, Satyam Computer (SAY) $14 ½, Scientific Games (SGM) $8 ½, Sonoco Products (SON) $28, Taiwan Semiconductor (TSM) $20, TJX $41, Urban Outfitters (URBN) $25, Wendy’s (WEN) $31, Whirlpool (WHR) $74, Watsco (WSO) $14 ½.

NASDAQ (47)
ABC Bancorp (ABCB) $14, Arkansas Best (ABFS) $31, AFC Enterprises (AFCE) $29, Area Bancshares (AREA) $19 ½, Borland Software (BORL) $18, Bisys (BSYS) $66, Cheesecake Factory (CAKE) $37, Concurrent Computer (CCUR) $17 ½, CDW Computers (CDWC) $59, Cephalon (CEPH) $78, Connective Thera (CNCT) $14 ½, Cyberonics (CYBX) $29, Casella Waste (CWST) $15, Drexler Tech (DRXR) $25, Dynacq (DYII) $29, Fuel-Tech (FTEK) $6 ½, Freemarkets (FMKT) $27, Foamex (FMXI) $9 ½, FPIC Insurance (FPIC) $15 ½, Freds (FRED) $43, Global Sports (GSPT) $23, GK Svcs (GKSRA) $35, Genesee & Wyoming (GNWR) $33, Glycogensys (GLGS) $2 ¼, Global Imaging (GISX) $21, Heartland Express (HTLD) $31, Igate Corp (IGTE) $4 ¾, Integrated Circuit Systems (ICST) $27, JB Hunt Transports (JBHT) $27, Kulicke & Soffa (KLIC) $20, MTR Gaming (MNTG) $17, Numerical Technology (NMTC) $39, Ohio Casualty (OCAS) $16 ½, O2Micro Intl (OIIM) Overture (OVER) $43, PEC Solutions (PECS) $44, Progress Software (PRGS) $19 ½, Premiere Tech (PTEK) $4 ¼, Roadway Express (ROAD) $41, Ross Stores (ROST) $35, Sonosight (SONO) $38, Total Entertainment Restaurant (TENT) $6, Trident Microsys (TRID) $8, Ustarcom (UTSI) $35, USA Truck (USAK) $12 ½, United Stationers (USTR) $36.

7. Using Buying Climaxes for Industry Analysis

In December 2001, we noted that the buying climaxes included a number of Restaurant stocks. When a number of stocks in a particular group have a lot of buying or selling climaxes, it is worth watching the group for a potential change in sentiment. They included the following stocks:

Applebee’s (APPB) $35.20 (20 P/E 0.2% yield). Earlier, we had reported a negative broadening top formation and suggested stock might get up to $36 which it did on chart. Low here was $13 1/2 in 7/00. There are 36 insider sellers, no buyers.

Cheesecake Factory (CAKE) $35.49 (46 P/E nil yield). This has been a good steady, up trending stock for a number of years, had a buying climax after $32 high in November 2000 and then went to down to $22 in March 2000. This was followed by another buying climax at $32 in July, 2000 followed by a drop to a low at $20 in September, 2000. CAKE had one insider buyer back last May, 2001 when the stock was relatively low and 7 sellers.

Darden Restaurants (DRI) $35.97 (21 P/E: 0.2% yield) has also been a good performer for a long time, selling as low as $7 in January, 1997. After moving to $23 with a Buying Climax at $19 1/2, the stock came down and had a selling climax at $14 1/2, made a low at $13 1/2 and then started its up move to the chart high at $37 and buying climax. There are six insider sellers here, no buyers.

Max & Ermas Restaurants (MAXE) $15.55 is one we do not follow closely. It does have a 29 P/E and no yield.

Ryans Family Steak House (RYAN) $20.89 (14 P/E: nil yield). RYAN bottomed with a Bear Trap after a $7 1/2 low in September, 2000. It would have a High Pole $19 1/2, but next sell signal here would be first in the up trend and stock might be good buy in $16-16 1/2 area. There are 14 insider sellers here, no buyers.

Tricon Global Restaurants (YUM) $50.02 (17 P/E, no yield) Stock is up from $24 low, where it previously had selling climaxes at $29 and $26. P/E was under 10 back then. This buying climax from a $53 top follows one in November, 2001 at $52 so it is showing signs of distribution. It would give a sell signal at $47 which would be the third in the up trend, typically the most dangerous. It has no insider buyers and 5 sellers.

8. Confirmation from the Industry Group Bullish % Indicator

Since December, 2001 Restaurant stocks have continued to rally but now (mid-Feb, 2002), the Bullish % indicator for the Restaurant industry group is overbought at 74%. Previous action for this indicator was a top at 70% last August, 2001 followed by a drop to 40% with a sell signal in September, 2001.

Historic peaks were as follows:

June 1987 70%
June 1990 76%
January 1993 70%
January 1994 72%

The above evidence from both the frequency of climaxes in this group together with the overbought nature of the bullish % indicator suggest that we will be getting a top in the near future.