SMarketModels

data revealing hidden market information that could otherwise not be gleaned from a typical bar chart

8.5 Data Sources

For more information on the indicators in the Sentiment Model, refer to the sources in Table 8.11.

Table 8.11. Data Sources

Data

Volatility Index (VIX) Put/Call Ratio New Highs

New Lows Arms Index (TRIM) Bullish Consensus Short Sales Ratio

Source

Chicago Board Options Exchange Chicago Board Options Exchange

New York Stock Exchange New York Stock Exchange New York Stock Exchange Market Vane

NYSE Members Report

9 Tools ofthe Trade

Generally he who first occupies the field of battle to await the enemy will be rested; He who comes later and hastens into battle will be weary.

Sun-tzu, The Art of Warfare

Preparation for the next trading day begins at the closing bell, and by time the opening bell rings, traders work should be done for the rest of the day, except for the simple matter of executing the trades when alerts are received. During the trading day. priorities are order entry and money management instead oflrade selection - an exercise of managing risk rattier than creating it.

Some traders like to come in with a blank slate. If one chooses to trade this way, then discipline is a greater factor. As long as risk is managed, discretionary trading can be equally rewarding. As an analogy, system trading is like listening to classical music, interrupted occasionally by the counterpoint of subtle mouse clicks. In contrast, discretionary trading is like blasting rock music, punctuated by the sensory overload oftick data.

Even with discretionary trading, preparation is the key to success. Knowing when to pull the Irigger and to wait patiently for tiie right trade will eliminate some of the angst. The only requirement is tiiat tiie Irade must have a catalyst. Just because a Level II screen is showing upward momentum does not mean the Irade has a catalyst, but combine that momentum witii a catalyst and there is a basis forthe trade.

Consider the example of a stock in a strong downtrend that a trader knows is heavily shorted. In tiie middle ofthe day, tiie company aimounces a big stock buyback. The trader sees the news release and buys the stock instantaneously. The trader docs not hesitate because the catalyst is apparent. The foundation of

success here is the traders knowledge ofthe stocks fundamental and technical condition, an awareness of a catalyst that can change it, and a definition ofrisk (e.g., exit the position if the low of the previous five minute bar is broken).

9.1 Tyco Case Study

Consider the case ofTyco (TYC:NYSE) in Figure 9.1. CK-er a period ofthree weeks in eady 2002, the stock dropped in price from the mid-fifties to the low thirties because of accounting concerns. Furthermore, the comparJ artoounced that it would spin off some ofits operating companies rather than continue its acquisition strategy (note the gap up and reversal on Januarj ").

Figure 9.1. Tyco Daily Chart

OnJanuarj 30, the stock gapped down over three points. Given the possibihty of a positive catalj-st, the stock was placed on the watch hst. After the stock gapped down, it drifted downwards and languished the rest offhe moming

At 12:08 pm, Tyco issued a press release that the CEO and CFO would be purchasing 500,000 shares apiece with their own money. The timing offhe release could not have been more perfect Wldle some traders were eating lunch, others were ha\ing it eaten because the price spiked fiom 28 to 30 in a matter of minutes (Figure 9.2) and then climbed the rest offhe afternoon to close near 35. When a stock has declined as much as Tyco, only a positive catalyst is required to reverse the stocks direction in a matter of hoiu-s. We recommend that the trader set up awalch fo/ofstocks in the news to be alerted to potential reversal catalvsts.

111 ll....... . IF-,. \ .. i

.llll......<;..i.l..ii(:< u.m

. ,1.......1 I., < >lu.. S...... .1.11,hi)- Ml, liji I I >»i,,.I.i . tIk

......>l ll.............-.......iMi I..... I mi.lii.t..i >v.......

Figure 9.2. Jyco hitraday Chart

Some traders prefer not to wich or monitor the news at all, but the trader gains an edge in situations hke Tyco-these situations do not occur often, but the key is that the news must be a reversal catalyst, hi contrast, ifa company reports bad news while its stockremains in a downtrend, fighting the tape is futile.

9.2 Preparation

To prepare for the next trading day, we take the following steps at the end ofthe current trading day. Ah offhe work for the next trading day is done the prewous night. For all intents and purposes, the trader walks in the following moming, gets trading alerts, and executes orders. The time between 9:30 am and 4:00 pm should be the most boring hours ofa traders life.

1. After the close, download todays closing prices.

2. Run the scanning software to generate orders and new alerts.

3. Review the open positions.

4. Analyze the charts with new alerts

5. Assign priorities to charts for trading the next day.

6. Enter the symbols and/or alerts mtu the trading software

E-entually, the traders routine should become fairly rigid. Occasionally, panic will strike when the trader is trjing to liquidate a large position in an illiquid stock (hint), or when the Federal Reserve Board decides to imleash a surprise interest rate cut on the market in the middle ofthe day. The old saving "hours of boredom intermpted by moments of panic" appUes to the trader, which is equally appreciated by sports fishermen and the guy holding the down marker at aprofessional football game.

9.2.1 Software

The key to establishing a trading routine is to automate as much of the process as possible, and that means investing in software and hardware. The minimum requirements are: one computer, two data feeds (for redundancy), two monitors (one for trading and one for analysis), and one technical analj-sis product.

We use the TradeStation 2000i software for nightly scans. The new version of the produtt, TradeStation 6, does not include end-of-day scans on a portfolio of stocks, so after the nightly scan, we take the candidates forthe next day from version 2000i and track them real-time in the version 6 produd; . . the Acme software runs on both versions). A software alternative to TradeStation is the MetaStock Professional product.

Other products such as FirstAlert® are designed to scan the whole market, notjust aportfoUo of stocks. The ad-antage ofreal-time scanning programs is that their formula languages are optimized for speed, creating djmamic lists of stocks on a tick-by-tick basis. The disad-aniage is that a real-time scanner can generate sensory overioad, and the trader may be tempted to jump ftom one stod: to anotiier.

Other traders prefer and need the discipline of a program hke TradeStation because everything can be automated: the number of shares to trade, the exact entrj and exit points displayed on the charts, and even order execution itself. Further, one can code an idea quickly and \iew the results ofback testing.

The most important issue is the "cleanliness" ofthe data, which is affected by bad ticks. When analyzing performance, rewew each trade to ensure that the results are not flawed, looking for unusually good or bad outUer trades. Just a single erroneous trade can skew the results. Finally, reahze that no one produd; offers the ultimate trading machine. Each programs has its strengths and week-nesses, and no program is bug-ftee. Trading software is very difficult to write

properly, so the vendor must be able to respond to problems in real-time with a competent technical support department.

9.3 A Trading Day

Think ofthis chapter as a "real time" chapler. Today is Monday, Februarj 25, 2002, and we are preparing for Tuesdays trading. The beautj ofthis experiment is that we have no idea what is going to happen with our stock selections, so at the end ofthis trading week, we are going to appear as either Adam Vinatieri or Scoft Norwood-

Table 9.1. Nightly Download

The first step is to download the closing prices, as shown in Table 9.1. After the download. Table 9.2 shows aU ofthe open positions (17) in the database. On given night, the number of open positions will -arj ftom ten to thittj for a database of approximataly 500 stocks. Note that ele-en of the positions were opened on today, Februarj 25, and that nine ofthe signals were long entries. Within the context ofthe market, this behavior is logical because the Nasdaq Composite Index hadjust completed a five-day downtrend, reversing to close above the open on Februarj 22nd, as shown in Figure 9.3.

On 2(NH. Ilic l44l.-i;il li.-siiw . ... ih<- (. . 1 ti..m .... ..,„,. „1 „ , .1.,..

.>.<-.. Wl ...I li.lMMl 111 ll.lli .111 I ..li.,vv./.-.....In - -

[lllu- miJ<ll(<.l fix- A:n. stilJ •VI-.....si.i.li- iL.V I"", svsli-...

ILrloi...... Illllll > IK 1>1 (1.1 4V.ll. MV.-.I M......I.

s.,,.,, It..wl WW......... Ni-,v -..,,l,...l I......... ,1,1 1.........,

>v.lli I..... .<......I-. .1.11........, ... III. V..........<• ...... Il<»vl WV

,...1 1.1 1.1