CISCO


CISCO Futures

1-303-306-1521 1-800 800 7227 Fax 1-303-368-9449
Internet: http//www.cisco-futures.com
Email dljones@cisco-futures.com



Trading Opportunity: The Figure of Merit
Dow Index versus S&P

Donald L. Jones©
March 13, 2004

Traders are always searching for the ideal, or at least, a winning, trading model. Experience shows that given an exact model saying buy here, sell there, one person may win while the next loses. Why is this? Likely the differing results arise because a model is inevitably in two parts. Part one is the calculational piece that spits out buys and sells. Part two is the input from the trader, all depending on the trader's market knowledge and attitude toward risk and treatment of gains. Trading Model Development

This article is about evaluating your own trading using trading risk and trading opportunity. No, it is not a study of diversification or market timing. It is about knowing what you are doing, putting yourself in a position to win. You can trade blindly or you can understand how the approach you are following, your 'model', fits the market(s) you are trading. You can generate a Figure of Merit for your model. That measure is the ratio of potential (opportunity) to the risk (P/R) on a trade by trade basis. With a little work you can calculate the P/R ratio from the trades you have completed. We illustrate the principle with the CISCO Basic Model, comparing the Dow Index to the S&P.

Model/Market Evaluation: The Concept of Figure of Merit
You can evaluate any model you use by taking the entry price and checking what the market did after entry. You entered and the market immediately went against you. You had little opportunity to make a profit. Or, you entered and the market went your way. The market offered you an opportunity to make a profit. You probably did not exit at the top (or bottom), but if you had you would have made the maximum profit possible. This maximum is what we call potential or opportunity. Virtually every trade has opportunity. Every trade carries risk. For most cases risk is the stop you placed on entry. The ratio of opportunity divided by risk is the Figure of Merit (FOM). FOM puts you in a position to evaluate both your model and yourself. If your trade is in a certain market that typically offers little opportunity, you probably should find a different market to trade. If the average potential (opportunity) is small compared to the average risk, again you have a decision to make. If opportunity is adequate and your returns are not, you need to examine the "you" aspect of your trading.

Markets change. Opportunity, the average maximum profit one could make from a trade, may change over time. Today's marvelous model may be tomorrow's dog. If you regularly measure the opportunity of your completed trades, the FOM, you can tell when your market is going into a funk.

Using CISCO Basic Model for Opportunity and Risk
Opportunity for day traders is measured from entry. It is simply the best return possible after entry, tracked to the close or stop-out, whichever comes first. Measuring opportunity requires first setting up a model for entry. We use the CISCO "Basic Model" to identify a balanced market breakout. Basic Model locates the price range of a balance and uses the upper and lower limits of the range as the breakout entry points.
You can read about balanced markets at: Overlay Demand Curve Background.

Basic Model also gives us a measure of risk. Comparing the Overlay to a quasi-bell shaped curve, we can think of a breakout as marking the change in distribution from balance to trend. When a breakout fails, with price coming back into the original distribution, the failure is confirmed when price penetrates back about one-eighth of the original upper limit - lower limit range. An eighth of the original balance is called an Octant. One octant then, is the risk. After entry, risk may be used as a trailing stop.

The breakout Basic Model is very simple. Go long (short) at a price one tick past the upper (lower) limit. Track the trade with a trailing stop of one Octant.

Example of an Overlay Demand Curve
The Overlay in Figure 1 is a single-distribution, balanced market for a single day. Ten days of Market Profiles of the December 2003 DJ future from October 8 through October 21 are linearly added, making the composite Figure 1. To learn how TPOs are generated by the Market Profile see: Market Profiles Background.

The upper limit of the distribution is 98000 and the lower limit is 95900. Risk, the Octants, are at 97750 and 96150, respectively. For tomorrow, the market of October 22, you know where the breakouts may occur and what the risk is if a trade is initiated.



   Ten Day Overlay for a Balanced DJIA Market

TPO VOLUME OVERLAY AND PRICE ROTATION PROFILE
DEC 03 DJIA (CBOT)  DAY      
10 08 03 TO 10 21 03   (Ten trading days)

 PRICE DYS  L/F ROT PROFILE *  TPOS TPO VOL OVERLAY *
 
 98400  1         5               1 X
 98350  1         5               3 XX
 98300  1         5               3 XX
 98250  1         5               3 XX
 98200  1         5               2 XX
 98150  1         5               2 XX
 98100  1         5               2 XX
 98050  1         5               2 XX
 98000  2         56              5 XXXX <== Upper Limit
 97950  3         567             7 XXXXX
 97900  3         567             9 XXXXXXX
 97850  4         4567           12 XXXXXXXXX
 97800  4         4567           14 XXXXXXXXXXX
 97750  5    9    45679          18 XXXXXXXXXXXXXX <== Upper Octant
 97700  5    9    45679          21 XXXXXXXXXXXXXXXX
 97650  5    9    45679          24 XXXXXXXXXXXXXXXXXX
 97600  6    9    456789         21 XXXXXXXXXXXXXXXX
 97550  6    9    456789         22 XXXXXXXXXXXXXXXXX
 97500  7    9    3456789        27 XXXXXXXXXXXXXXXXXXXXX
 97450  7    9    3456789        28 XXXXXXXXXXXXXXXXXXXXXX
 97400  7    9    3456789        33 XXXXXXXXXXXXXXXXXXXXXXXXX
 97350  8    9    13456789       38 XXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97300  7    9    1346789        46 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97250  7    9    1346789        52 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97200  7    9    1346789        51 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97150  7    9    1346789        48 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97100  7    9    1346789        42 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 97050  7    9    1346789        33 XXXXXXXXXXXXXXXXXXXXXXXXX
 97000  7    9    1234789        22 XXXXXXXXXXXXXXXXX
 96950  5         12378          15 XXXXXXXXXXXX
 96900  4         1278           14 XXXXXXXXXXX
 96850  4         1278           11 XXXXXXXX
 96800  4         1278            8 XXXXXX
 96750  3         127             9 XXXXXXX
 96700  2         12              8 XXXXXX
 96650  2         12             10 XXXXXXXX
 96600  2         12             13 XXXXXXXXXX
 96550  2         12             13 XXXXXXXXXX
 96500  2         12             12 XXXXXXXXX
 96450  3   Z     Z12            16 XXXXXXXXXXXX
 96400  3   Z     Z12            11 XXXXXXXX
 96350  3   Z     Z12             8 XXXXXX
 96300  3   Z     Z12             5 XXXX
 96250  3   Z     Z12             7 XXXXX
 96200  2   Z     Z1              7 XXXXX
 96150  2   Z     Z1             10 XXXXXXXX <== Lower Octant
 96100  1   Z     Z              11 XXXXXXXX
 96050  1   Z     Z              13 XXXXXXXXXX
 96000  1   Z     Z              13 XXXXXXXXXX
 95950  1   Z     Z               8 XXXXXX
 95900  1   Z     Z               3 XX  <==  Lower Limit
 95850  1   Z     Z               1 X
 95800  1   Z     Z               1 X
 95750  1   Z     Z               1 X
 95700  1   Z     Z               1 X
 95650  1   Z     Z               1 X

Figure 1. Overlay Demand Curve: Ten days, DJ, October 15 - 21, 2003.
Price is descending in the first column, the X's on the right are volume, peaking at 97250. The column 'ROT PROFILE' identifies the days' prices, day 9 (Oct 21) ranged from 97750 to 97000, day Z, ten days back (Oct 8), was 96450 to 95650. The next column, 'TPOS', is a count of the Market Profile TPOs that went into the Overlay. 'TPO Volume Overlay' column is a visual depiction of the TPOs in the TPOS column. Upper and Lower Limits are the limits of this Overlay distribution. Upper and Lower Octants are the risk.

A Trade on October 22
At the close of October 21 the limits are 98000 and 95900, risk is about 260.
On Oct 22, price traded below 95900, with a short entry at 95890. After the entry, price went as low as 95550. The Potential, the maximum this short trade could have possibly returned after entry, is 95890 - 95550 or 340 points. 340 points is the Opportunity. Opportunity is the return the perfect trader would attain. Risk is one-eighth of the limit-to-limit range or about 260. For this day, the potential-to-reward ratio (Figure of Merit), is 340/262 or a modest 1.3.

Using the Basic Model, CISCO's Advice Engine daily surveys the entire futures market and prepares a list of possible trades. The only criterion for inclusion in the Advice Engine report is that the market be in balance. Advice Engine examines all the futures on the day's list. The ones deemed most probable to be profitable are placed on a 'Select' list.

Research
All traders have seen trading models offered for sale. These are effectively black boxes, generally with a number of conditions. It is difficult to do your own trading research with such information. That is not the case with the Advice Engine tables. Entry is on breakout from a 10 day Overlay. Risk is the Octant. Overlays are posted at the close of each day and checked for breakout the following trading day. If there was a breakout, the Potential opportunity is posted with the results. These data are not open to discussion. The market either broke out or it did not. If it did, the Potential is directly measureable and the risk is known from the balance limits.

Your model is doubtless not the CISCO Basic Model. But the principles involved the the Figure of Merit concept are the same for all models. The CISCO Select Tables can be your test bed. The data is available. Use that data to familiarize yourself with application of the principles and then turn to your own data.

Applying the Select Data Tables to DJ and SP
We have seen that opportunity/risk measurements can find markets to trade and markets to reject. We can test ourselves against the ideal. Currently, there are about three years of history in the Advice Engine Select database. These data can be used for all sorts of studies. Mentioned so far are the single future study, i.e. the one you trade. Another question might be which of two futures offer the better trading situation? We take that up in the study in Figure 2, comparing the DJ and SP.


                     Opportunity/Risk Table: DJ vs SP

  Del    Rsk-DJ  Rsk-SP   Potl-DJ  Potl_SP  P/R-DJ  P/R-SP  #Tds-DJ  #Tds-SP
 Mar 01     369    1269      1009     2265    2.85    1.78       19       16
 Jun 01     323    1246      1043     2775    3.23    2.65       20       19
 Sep 01     274    1154      1054     2375    3.85    2.64       15       24
 Dec 01     378    1121       612     1786    1.62    1.80       19       15
 Mar 02     342    1260       637     1583    1.86    1.52        4        3
 Jun 02     251     884       348     1358    1.39    1.86        6        3
 Sep 02     387    1434       980     3127    2.53    2.58        7        6
 Dec 02     391    1222       879     1990    2.25    2.26       17       10
 Mar 03     307    1026       527     1505    1.72    1.97       15       17
 Jun 03     307     981       958     2658    3.12    3.19        6        3
 Sep 03     270     964       708     1548    2.62    2.05       16       17
 Dec 03     186     664       357     1110    1.92    2.25       22       25

 Average    315    1101       759     2006    2.41    1.82     13.8     13.2

Figure 2. Figure of Merit for DJ and SP.
Definition of columns: Del is the delivery month, Rsk is the average risk for that delivery month, Potl is the average opportunity for the delivery, Figure of Merit, P/R, is the Opportunity-to-risk ratio; and #Tds counts the number of trades for that delivery.

Note: Conversion to $: DJ x 1, SP x 2.5 (both full contracts). The ratios of of Potential-to-Risk, Figures of Marit, are independent of the dollar conversion.

Results
An average Figure of Merit for DJ of 2.41 versus SP of 1.82 shows DJ to be 32 percent better on an opportunity-to-risk basis [(2.41 - 1.82)/1.82 x 100]. On the face of this measure, a trader has a much better play in the DJ.

There is an apparent correlation between the average P/R ratio and the number of trades in that quarter. Dow futures showed an opportunity-to-risk ratio of 3.85 for the September 2001 quarter and a mere 1.39 for the June 2002 quarter. No matter how well you trade, the June 2002 quarter offered only 36 percent of the opportunity of the previous September.

As the market falters, opportunity tends to falter as well. Remember, these data points are for a quarter year, averaged. If one searched the individual trade by trade data, the tip-off to avoid that market (or treat it differently) would appear within a short while into the dull period. The dull period is characterized by fewer trades and poorer opportunity-to-risk ratios for those trades that do occur. So, if you see trading possibilities becoming rare and then the opportunity diminishes, beware!

Of the results, by far the most important is that you, the trader, can have an unbiased set of data against which to develop and evaluate your own trading model. There is nothing mysterious about opportunity-after-entry and you certainly know your risk. So you can continuously rate your trading model (is it generating adequate opportunity?) and your trading (are you converting enough of the potential to your bottom line?).

Decision Time: Which to Trade: DJ or SP?
Trading decisions are often qualitative and personal. The data in Figure 2 shows approximately the same number of trading opportunities for each of DJ and SP. The three year study in Figure 2 identifies the slow periods, e.g. December 2001 through September 2002, to be about the same for both.

An important factor is the risk capital required. DJ takes only about one-third that of SP. The SP Potential is larger, but the Figure of Merit favors DJ.

So which would you choose? Whichever it is, you now have some quantitative comparisons to help your decision!

Additional Trading Information on the CISCO site

CISCO Home Page: http://www.cisco-futures.com/
Market Profile: Market Profiles Background
Overlay Demand Curves: Overlay Demand Curve Background
Liquidity Data Bank: Liquidity Data Background
BuySell Data (CBOT): Liquidity Data BuySell Background
Short HomeStudy Course: ShortHomestudy Page
Long HomeStudy Course: Homestudy Page
Trader Control Package: Swing Trader Package(TCP) Page
Advice Engine: Advice Engine Background
Auction Market Theory: Auction Market Theory
Trading Model Dev.: Trading Model Development
Day Trading Supp/Res: Day Trading Support and Resistance
Volatility and Stops: Volatility and Stops for the DayTrader



About Donald Jones and CISCO
Don Jones has been researching and trading the futures markets for over 30 years. Along the way he initiated the use of tick data for Market Profiles in 1987, created the Overlay Demand Curve to find market condition (1988), published maybe the first ever study showing that markets were not efficient (1973), developed a sophisticated Market Profile generator for day traders (2000), proposed a 'current' volatility for option traders (2003) and more recently has continued work on Auction Market Theory first published in 2002. CISCO maintains Liquidity Data Bank history dating back to 1987 and Market Profile history to the same date.

CISCO developed early trader training courses based on auction market principles in 1989, upgrading them from ongoing research to the current day. The Short Course (5 months) emphasizes a combination of market condition and Market Profile information that is of most interest to daytraders.
Long Course (12 months) is more oriented to the needs of the swing or position trader.

Both courses promote learning through doing, one written lesson per month accompanied by daily trading practice.