Tom, thank you for your response. I believe that you post your NASDAQ Intermediate Index value daily on the SI Public Charts List. Questions:

(1) Do you also post the NYSE Intermediate Index value in public domain?

(2) How do you aggregate NYSE and NASDAQ Index values into a single Composite NYSE/NASDAQ Index value?

(3) Is the determination of the 15 indicator value in the NASDAQ Intermediate Index completely automatic using certain formulas or somewhat subjective using visual assessment?

(4) Are the results using the Composite Index superior to using the NASDAQ Index alone when the top Sectors involve only or mostly tech sectors?

(5) Are the results of the system using the top Sectors superior to using only the 2x Profunds NDX and SPX/RUT index funds? (the reason for asking this question is the added leverage when using 2x index funds vs 1.5x sector funds which somewhat makes up for the greater volatility of sector funds)

George

Tom, you have an interesting system. I am wondering if you have done any backtesting of your system to determine its expected performance during bull and bear markets and also its maximum drawdowns as a percentage of the peak account values. Here are my thoughts on your system.

General points:

(1) It's a trend following system using relatively infrequent switches from long to cash/short based on the Intermediate Composite Model NYSE/NASDAQ Index (ICM Index) reading. It is designed to capture longer term uptrends and avoid longer term downtrends.

(2) The system uses a reasonable level of leverage (1.5 x Profunds sector funds) and a moderate degree of diversification (typically, various correlated tech sectors with greatest momentum)

(3) The system is fully mechanical since daily switches are completely determined by the ICM Index-based long/cash/short signals and by the Sector selections using the 100 day Relative Strength Composite values (RSC values).

Specific points:

(1) The first key aspect is the quality of the long/cash/short signals. Some systems are anticipatory ("early" entries) and some systems are reactive ("late" entries). It will be interesting to see into which category your system falls based on a reasonable track record.

(2) The second key aspect is the quality of the position sizing algorithm. Your system appears to employ automatic gradual position increases in 25% increments ("pyramiding") which should be a sound approach. However, this also depends on the quality of the long/cash/short signals and the resulting average gain from the time the ICM Index moves above 0 until the time the ICM Index moves below 0.

(3) The Sector selection using RSC values appears somewhat less important since highest momentum sectors will be usually various tech sectors which are both highly correlated to each other and to NDX. Yet, it should provide some additional gains due to relative outperformance of highest momentum sectors.

(4) The system assumes that the 100% aggregate long position size will be maintained when the ICM Index is above .20. There is some question whether this is the optimal strategy given Kelly's formula for fixed-fraction betting and given the related probabilities of gambler's ruin outcome. Moreover, the 1.5x leverage translates the 100% aggregate position into a 150% aggregate position in terms of the total account size.

George

2/1995

Hi Hul, I appreciate your very interesting info. I wonder if you still look at QQQ historical intraday patterns to make your trading decisions. Also, what does your position size depend on?

I haven't recently found too many traders on the threads who can consistently gain even 6% a month (100% a year) for more than a year (of course many did it easily during the 10/1998-3/2000 period). The best traders with at least a partially verified record that I encountered recently are:

(1) NDX funds/QQQ: ajtj (Profunds), Huluriasquias (Rydex), positiontrader (Profunds), wahz (QQQ)

(2) Options/futures: ajtj (options), TradeHard (futures), WinLoseOrDraw (options, futures)

But even among those traders above, very few if any have a record of average compounded 10%+ returns for the past 12 months (even though they may have achieved it for shorter periods of time). I also saw a few other traders on SI claiming 10%+ a month returns in the past and the actual record in one instance verifying the claim.

NoiseBox includes intelligent position sizing plus it is intraday which enhances the probability of profitable trades. So, it should be more profitable than a similar system based on eod closes simply because of higher frequency of trades.

On the other hand, Huluriasquias claimed returns of 7-8% a month compounded for the past 7 months earlier this year using Rydex funds (I believe) and eod trades and positiontrader (Marc) has had returns exceeding 13% a month compounded for the past 10 months using Profunds and eod trades.

So, it is still not clear to me whether 10-12+% a month is achievable and whether eod trading is able to achieve it. Of course, even 6% a month means 100% a year which is surely a superior return.

Mechanical systems are too simplistic in most cases:
(1) There are multiple factors that need to considered in trading decisions with appropriate dynamic weightings.
(2) Markets change over time so systems based on data mining over a restricted time period are of limited reliability.
(3) Position sizing may not be considered in many mechanical systems.
(4) Diversification may not be considered in many mechanical systems.

I don't know the answer either but I will certainly try to find out through my trading. Profunds funds are an ideal tool for that IMO. The goal in any case is to approach the theoretical upper performance limit subject to acceptable risk level (I still estimate this limit at 20% using Profunds 2x funds). I might consider the risk level acceptable if, say, maximum drawdowns do not exceed 10-15% of the total account value. I have to see in practice if this risk level is achievable and compatible with high returns.

WLD, I am reading several books dealing with issues of market risk, bubbles, trading, etc. as well as the academic literature on market inefficiencies. They are all very informative as they deal with the pertinent issues of trading systems, position sizing/leverage, and also emotional/psychology aspects.

Surely risk cannot be eliminated but it can be minimized/optimized. While my preferred approach is basically momentum investing using leverage, hedging and diversification are two techniques that I may use more extensively in the future to manage risk.

Lisa, I simply meant that, at best, T/A predicts the up or down market moves with a high probability but not certainty. For example, according to T/A (not to mention F/A) the market can be extremely oversold and yet still keep going down (Sep 2001) and it can likewise be extremely overbought and still keep going up (late 1999). Or, there can be a single day with an extreme market movement down (Oct 1987) or up (Jan 2001).

So, it is not optimal to bet 100% of the account based on T/A predictions since there is always a non-negligible probability that the oversold/overbought conditions will be continue. That's why money management (i.e., positon sizing) is the crucial aspect of a successful system. This is explicitly embedded in the Turtle system of Richard Dennis (who earned $200,000,000 through trading) and also emphasized by Ed Seykota (who had 100% average annual returns for 12 consecutive years and developed the first computerized trading system for a major firm).

Now, there are some tools in probability theory that are of help in estimating the correct position sizes (Kelly's formula for the fixed-fraction betting system) and in calculating the chances of "blowing up" that is losing 100% of the account (solution formula for the gambler's ruin problem). These are a bit too simplistic for the stock market case but still provide important insights.

All this becomes even more important when using leverage (margin, Profunds 2x leveraged funds, options, futures). Leverage is absolutely crucial in order to achieve average returns of, say, 100+% a year as many top traders have done (trading futures, options, commodities, etc.) And yet, without disciplined money management, leverage exacerbates that much more the risk of "blowing up" the account. Incidentally, I am reading now Lowenstein's book on Long Term Capital Management where some these issues are brought up.

ATTENTION: Modified System will be used for trading from today.

(1) As far as the details of the Modified System, it will soon be evident from my trades.
Generally speaking, there will be more trades in the direction of the trend and fewer against the trend.
Also, modified position sizing will be utilized.
Finally, trading vehicles will be more diversified.

(2) Modified System might be considered an euphemism for Malfunctioning System. However, the reason for the change is that I have carefully studied what conditions my system must necessarily satisfy in order to achieve 10+% monthly returns.

The first one is: avoid lost opportunities connected to large market moves (in either direction).
The second one is: avoid excessive losses.
The third one is: limit position risk to avoid either being stopped out of positions or incurring large losses.

(3) The Modified System is still not mechanical as I have not found satisfactory mechanical rules to apply in order to achieve optimal entries and exits.

(4) T/A unfortunately does not provide satisfactory answers to many questions regarding system signals to make them very reliable. So, as Ed Seykota, Richard Dennis, and others have noted, the answer must lie in optimizing position sizing/money management aspects of the system.

George

For the period 1/1/2000-8/14/2003 which is 43.5 months, the average monthly sum of absolute daily NDX % changes is 49.11 which I rounded to 50% a month. However, for the period 7/1/2003-8/14/2003 which 1.5 months, this sum is only 36.5% or 24.3% a month which is roughly half of 50%. Given this, the expected returns from trading during the past 1.5 months might be proportionately lower due to this lower volatility.

As far my lack of trades for one week, a few points. First, that 2% daily NDX change (actually 2.36% a day for 2000-2003) is not uniformly distributed so gains are larger when NDX turns from very overbought or very oversold. Second, the correct trade was long entry late last week and early this week and less so now. Third, after I exited longs late last week due to lack of clarity on my system signals I spent a few days reevaluating my system and developed substantial improvements I mentioned earlier.

To reiterate, there is no need to trade every day. It is sufficient to trade when either the win probability or the potential gain is the highest. So, say, 10 trades a month timed for the most suitable setups (i.e., most reliable signals) will likely provide the great majority of theoretical gains possible in a given month.

WLD, let me explain the 10% number...

assume that we have these daily % changes in NDX during the 10 trading days next month:

 
 +2% -2% +3% -2% +1% +5% -2% +1% -3% +4%   
 

the corresponding absolute daily % changes in NDX during the 10 trading days next month will be:

 
 +2% +2% +3% +2% +1% +5% +2% +1% +3% +4%   
 

the sum of these 10 absolute daily % changes is 25%...so, similarly, over 20 trading days the sum of the 20 absolute daily % changes would be 50%...

my Excel spreadsheet shows that 50% was the average sum of the 20-21 absolute daily % changes over a typical month since 1/2000 till now...

so, if one has only winning trades, one can capture the full 50% of the sum of absolute daily % changes over a typical month...and in Profunds, this would mean capturing the full 100% of these absolute daily changes (of course compounded 50% would produce more than a 50% return)...

now, one will have some winning trades and some losing trades...assuming one has 60% of daily winning trades and 40% of daily losing trades (which is probably realistic for a top trader) and assuming that winning and losing daily trades have on average the same daily percentage changes (which is probably realistic as well) one can expect to accumulate

.6 (50%) + .4 (-50%) = 10%
of the 50% average sum of the 20-21 absolute daily % changes over a typical month...

using Profunds this doubles to 20% of 100% average sum of the 20-21 absolute daily % changes over a typical month

that's how I estimate 10% in NDX gains per month and 20% in Profunds gains per month...all subject to above assumptions on % of winning trades and on average sizes of winning versus losing trades...

it's possible that average winning trades could be larger than average losing trades in which case the % of winning trades could be just 50% and still produce reasonable gains...finally position sizes matter as well and, to simplify, I was assuming so far that all trades are done with equal position sizes...

in reality, one does looks at all trades even multi-day ones simply as single trades so that the calculations above would not correspond to typical record of trades unless one looks at the results day by day...

regardless of the particular computational approach though, I do get a ballpark figure of 10% in NDX and 20% in Profunds as a reasonable limit of what can be captured on a monthly basis...

I think what I meant is that when I misjudge the trend direction I should lower my exposure and, conversely, when I judge correctly the trend direction I should increase my exposure...I do not have an algorithm for position sizing only a general concept...the broad principle is to cut your losses short and ride your winners applied to the overall account performance...

Lisa, I like your comments and I know you are a very sharp person. You are right in everything you say. Here are my thoughts.

(1) There are three aspects of trading that have to be mastered to achieve high returns: system development, position/money management, and emotional discipline.

I am working on all of them simultaneously. You can see in my trades that I vary position sizes and I also vary the trading vehicles (NDX, SPX, US Bonds). I am very mindful of gambler's ruin problem which is more acute when trading leveraged funds like Profunds. That is also why the trades are based on a system which is not fully mechanical. In addition, to counter the gambler's ruin problem I am attempting to dynamically decrease my position sizes after unsuccessful trades and increase them after successful trades.

The emotional discipline is maintained by entering trades based on valid system signals and also by pausing trading and re-evaluating the system when the results are sub-par. Positive trading results increase my confidence in the system and therefore my emotional capability to trade based on system signals. Position sizing also plays a role in maintaining emotional discipline since lower position sizes reduce my risk after unsuccessful trades.

(2) In general, based on my readings and observations, I believe that markets are largely efficient so that above-average returns (adjusted for risk) are very difficult to achieve.

Yet, a number of individuals were able to achieve above-average returns historically, possibly due to "pockets of inefficiency" in markets. I am trying to determine through data analysis if above-average returns can be achieved and to what extent (say, 10% a month on NDX on average using EOD trading).

(3) My comment regarding institutionalizing my system needs to be further elucidated here because of possible (unintended) mercantilistic connotations.

The only interest I have as far as "institutionalizing" my system is to do what some of the most famous traders have done in the past. I intend to trade, if successful, until I grow my capital to a reasonably large size. At that point, and only at that point, I would consider starting a hedge fund of which my own money would constitute a large percentage. The idea is to continue trading using my system as usual for years but, at some point, accept large accounts for pooling with my own account in a hedge fund format.

In fact, out of curiosity at this point, I am reading up on hedge funds performance and strategies to assess performance levels and consistency expected in the hedge fund area. Interestingly, hedge funds make substantial efforts to develop proprietary trading strategies which have academic validity in terms of market anomalies (departures from market efficiency). Financial Analysts Journal is an accessible source of information on these issues.

Marc, in general I prefer high probability trades and right now I don't see any obvious ones.

Frankly, I am actually more and more reluctant to publicize my trades. On the one hand, when I publicize my trades I gain feedback from others and, more importantly, I feel peer pressure to do very well since my trades are public. However, I also believe that my system has a high potential to deliver exceptional returns. And so, I would like to maintain the proprietary nature of my system since I may institutionalize it down the road. That's why it is very possible that I will have to discontinue publicizing my trades after one or at most two additional months of 10%+ returns (if that happens, of course).

WLD, as I stated before, my goal is to capture the maximum possible percentage of that 50% average total NDX daily % changes per month. I am constantly analyzing NDX data to see what percentage of that 50% monthly total is achievable on average (consistently in the long run). It is really mainly a question of what's theoretically possible given the probabilistic behavior of daily NDX changes. An approximate answer to that question should imply a corresponding actual monthly trading return.

My conjecture is that capturing 10% a month of NDX daily changes is the theoretical limit in the long run. This 10% a month is equivalent to a net 20% of 50%
(simplifying, assume 60% of wins and 40% of losses, so
expected value= .60(50%) + .40(-50%) = 10%).

Capturing 10% a month in NDX translates into capturing 20% a month in Profunds (ignoring compounding effects). 20% a month equals about 9-fold increase per year or 800% a year. This 800%, I believe, is the theoretical limit for EOD NDX trading in Profunds. If I am correct about 20% a month as a theoretical limit, then 15-20% a month may be achievable in practice.

WLD, over the weekend I have substantially improved my system. I have added one new indicator, improved the use of the other indicators, diversified the system to other indices/sectors, and sharpened my position sizing/money management techniques.

According to the "improved" system, 50% of my trades since early June were either fully or partially faulty. The improved system should allow me to capture a reasonable percentage of the total of NDX daily percentage changes which is 50% a month on average (since 2000) or 100% using Profunds (the 50% number was obtained in Excel by adding absolute daily percentage changes in NDX over an average month or 20.85 trading days).

Right now I am waiting for a valid system signal to trade.

yep

thnx

TNX:
http://stockcharts.com/def/servlet/SC.web?c=$TNX,uu[r,a]maclynay[d19900101,20030731][pd20,2!f][J1...

Discography

http://www.led-zeppelin.com/disc.html

Video Vault

http://www.led-zeppelin.com/vv1.html

Photo Gallery

http://www.led-zeppelin.com/photogallery.html

Audio-Video Library

http://www.led-zeppelin.com/multimedia.html

Rock and Roll Hall of Fame Induction

http://www.rockhall.com/hof/inductee.asp?id=141

http://members.rogers.com/kastelco/m/LED_ZEP-STH.mp3

"A New Kind of Science" by Stephen Wolfram:
http://www.amazon.com/exec/obidos/tg/detail/-/1579550088/104-0338908-7709572?v=glance

Amazon.com
Physics and computer science genius Stephen Wolfram, whose Mathematica computer language launched a multimillion-dollar company, now sets his sights on a more daunting goal: understanding the universe. Wolfram lets the world see his work in A New Kind of Science, a gorgeous, 1,280-page tome more than a decade in the making. With patience, insight, and self-confidence to spare, Wolfram outlines a fundamental new way of modeling complex systems.
On the frontier of complexity science since he was a boy, Wolfram is a champion of cellular automata--256 "programs" governed by simple nonmathematical rules. He points out that even the most complex equations fail to accurately model biological systems, but the simplest cellular automata can produce results straight out of nature--tree branches, stream eddies, and leopard spots, for instance. The graphics in A New Kind of Science show striking resemblance to the patterns we see in nature every day.

Wolfram wrote the book in a distinct style meant to make it easy to read, even for nontechies; a basic familiarity with logic is helpful but not essential. Readers will find themselves swept away by the elegant simplicity of Wolfram's ideas and the accidental artistry of the cellular automaton models. Whether or not Wolfram's revolution ultimately gives us the keys to the universe, his new science is absolutely awe-inspiring. --Therese Littleton

From Library Journal
Galileo proclaimed that nature is written in the language of mathematics, but Wolfram would argue that it is written in the language of programs and, remarkably, simple ones at that. A scientific prodigy who earned a doctorate from Caltech at age 20, Wolfram became a Nobel-caliber researcher in the emerging field of complexity shortly thereafter only to abscond from academe and establish his own software company (which published this book). In secrecy, for over ten years, he experimented with computer graphics called cellular automata, which produce shaded images on grid patterns according to programmatic rules (973 images are reproduced here). Wolfram went on to discover that the same vastly complex images could be produced by even very simple sets of rules and argues here that dynamic and complex systems throughout nature are triggered by simple programs. Mathematical science can describe and in some cases predict phenomena but cannot truly explain why what happens happens. Underscoring his point that simplicity begets complexity, Wolfram wrote this book in mostly nontechnical language. Any informed, motivated reader can, with some effort, follow from chapter to chapter, but the work as a whole and its implications are probably understood fully by the author alone. Had this been written by a lesser scientist, many academics might have dismissed it as the work of a crank. Given its source, though, it will merit discussion for years to come. Essential for all academic libraries. [This tome is a surprise best seller on Amazon. Ed.] Gregg Sapp, Science Lib., SUNY at Alban.
- Gregg Sapp, Science Lib., SUNY at Albany

Book Description
This long-awaited work from one of the world's most respected scientists presents a series of dramatic discoveries never before made public. Starting from a collection of simple computer experiments---illustrated in the book by striking computer graphics---Wolfram shows how their unexpected results force a whole new way of looking at the operation of our universe.

Wolfram uses his approach to tackle a remarkable array of fundamental problems in science: from the origin of the Second Law of thermodynamics, to the development of complexity in biology, the computational limitations of mathematics, the possibility of a truly fundamental theory of physics, and the interplay between free will and determinism.

Written with exceptional clarity, and illustrated by more than a thousand original pictures, this seminal book allows scientists and non-scientists alike to participate in what promises to be a major intellectual revolution.

About the Author
Stephen Wolfram was born in London and educated at Eton, Oxford and Caltech. He received his PhD in theoretical physics in 1979 at the age of 20, and in the early 1980s made a series of discoveries which launched the field of complex systems research. Starting in 1986 he created Mathematica, the primary software system now used for technical computing worldwide, and the tool which made A New Kind of Science possible. Wolfram is the founder and CEO of Wolfram Research, Inc.---the world's leading technical software company.

so, all the Ph.D. traders are here...I like this group...

Marc, good thinking.

Marc, look at these market breadth indicators:

(1) Advance-Decline Issues/Volume:
http://stockcharts.com/candleglance?$NYAD,$NYHL,$NAAD,$NAHL,$AMAD,$AMHL

(2) McLellan Summation Indices (based on Advance/Decline data):
http://stockcharts.com/charts/indices/McSumNASD.html
http://stockcharts.com/charts/indices/McSumNYSE.html

(3) Bullish Percentages:
http://stockcharts.com/candleglance?$BPNYA,$BPCOMPQ,$BPOEX,$BPNDX,$BPSPX,$BPINDU,$BPDISC,$BPSTAP,$BP...

Marc, tomorrow will be more decisive if we end green today.

yes, I have a signal to go long...but my system has not been backtested for breadth indicators (advance/decline)...and they may have topped so the risk is pretty high if they did

BTK has been the leader

WLD, it won't be just a display of my market "prowess" but possibly it may indicate the maximum potential of T/A as far as superior market gains when trading EOD only...

my "magic box" will give signals at intervals that are likely to very much surprise you...but, modifications may still be required to fine-tune the system...I really wish there was more science/math/stat to it whereby one can achieve larger gains through more sophisticated/complicated systems...using neural nets, genetic algorithms, stochastic processes or what have you...but that seems NOT to be the case...

I have to say that I have finally discovered (after months if not years of effort) how limited is the usable info that T/A provides...generally speaking, T/A is like folk medicine, that is a scientifically suspect body of knowledge but with a few hidden gems with real scientific merit...

in fact, in the context of EOD trading, after my analysis I doubt that even a neural network can add value here...there is just not that much to work with really...it's just dynamic probability distributions that do look like random walk a lot in the aggregate but with a few "persistency" or autocorrelative features in a certain sense...after all it's not a pure lottery/casino...the human traders base their decisions on fundamentals and "sentiment" thus "stabilizing" the markets...in this context, I now also see the genius of Richard Dennis and his "Turtle Trading"...

of course, these few simple ideas I have can be ported into smaller time scales like hours, etc. to trade very successfully intraday...this, I suspect, is the reason for successes of all those famous futures traders...they also have very simple systems like mine but they can achieve possibly larger gains (??) due to higher frequency of up/down market moves...

TO ALL:

I have done a very intensive analysis of major aspects of T/A in recent weeks to determine "what works" in order to allow one to exceed market returns. Surprisingly, 90-95% of T/A DOES NOT WORK in any meaningful way IMO. The very few simple things "that work" cannot really be improved upon by any of the more sophisticated techniques of T/A due to inherent randomness of market moves.

Earlier, I had hoped that T/A can provide one with powerful techniques to "beat the market". Lately, I have been very apprehensive that perhaps NO T/A techniques really work and "market beating results" are entirely due to chance. Right now, I have concluded that it is indeed possible to "beat the market" using a few simple T/A ideas. In addition, achieved annual returns can be very large due to the fact that these simple T/A ideas allow one to substantially reduce losses while realizing a good portion of possible gains (both on the long and short side).

This conclusion is consistent with several known facts that I have been unable to reconcile until now:

(1) most academic research in finance found no statistical support for the proposition that market inefficiencies can be exploited to achieve above-market returns

(2) some more recent academic research in finance found that some simple T/A techniques do indeed result in above-market returns

(3) the original Turtle System traders achieved superior results following a well-thought out trading system


So from now on, I will implement these new insights in my trading. These ideas are a further refinement but also a major simplification of the ideas I arrived at early in June.

There is still one issue though that is presently unresolved. It is the optimal risk level one should assume. This issue has two overlapping aspects: leverage and position size. For example, if one uses NDX-based Profunds (or QQQ on full margin) with 2x leverage then a 100% position will result in percentage wins and losses twice that of NDX/QQQ. With futures, leverage may be as much as 5x or more. Normally, in these situations position sizes would proportionally decrease as leverage increases. And yet, higher leverage has the potential to boost the gains. For example, increased leverage may also be achieved by trading "high-volatility" stocks or options.

This issue is really mathematically related to the "Gambler's Ruin" problem which shows that the potential for a total loss of capital exists even when the trading system produces substantial gains on average. So, I intend to carefully monitor my system's performance from the standpoint of possible/likely drawdowns of capital given the system parameters (probability of a win=percentage of winning days, average daily percentage gain, and average daily percentage loss).

Gambler's Ruin problem and Practical Trading Applications:
http://www.math.usu.edu/~koebbe/GR/GamblersRuin/GamblersRuin.html

The Gambler's Ruin Simulator:

Let's assume that the gamblers start with $50 and will gamble until either $100 is reached or all the money is lost. Further assume that each gamble has a 0.6 probability of a $1 win and a 0.4 probability of a $1 loss. Let's assume that there are 100 gamblers and that they will stop after 250 plays (gambles).

To simulate 250 plays under these conditions enter the following numbers in the boxes below and press simulate:
50
100
0.6
100
250

This exercise will illustrate the chance of doubling one's money (from $50 to $100) without going bust (from $50 to $0) within 250 plays under the assumption that probability of a win is 0.6 and that win/loss is always $1 or about $1/$50=2% of the total amount. The results obtained for 100 gamblers indicate statistical distribution of the possible results.

This scenario is a good approximation of a trader trading a stock market index using technical analysis so that a win probability is 0.6 or 60% and the average daily gain/loss is about 2%. The simulation shows the chances of doubling one's money without going bust in 250 trading days (1 year).

Simulation results are very encouraging under these assumptions in that there is basically always a sizable gain after 250 plays, often reaching the desired $100. This indicates that 0.6 is a sufficient win probability to achieve good returns if one can keep single days wins/losses to about 2%. In fact, even 0.55 win probability is good enough for similar results. Similarly, changing the initial amount from $50 to $25 and thus win/loss percentage to about $1/$25=4% does not affect the results. Note also, that single day win/loss of 4% decreases to 2% as the amount grows above $50 to $100 (based on $1 single day win/loss).

So it appears, that the win probability is the KEY assumption and NOT the single day win/loss as a percentage of the total account. Therefore, as long as one can achieve win probability somewhat over 0.5 while keeping single day win/loss percentages balanced, the trading results should be superior.

The practical advice for every trader is to review one's past record in order to calculate
(1) percentage of winning days
(2) average 1 day percentage gain
(3) average 1 day percentage loss
to see whether
(1) > 0.55
(2) > (3).
If not, these two conditions should become the most important goals to be attained in trading.

T/A bibliography:
http://www.santafe.edu/~spyros/tabiblio.htm

Traditional arguments against T/A:
http://www.wwnorton.com/catalog/spring00/malkiel1.htm
http://www.wallstraits.com/lynch/quote1.html
http://www.usnews.com/usnews/biztech/articles/030616/16risky.htm
http://www.dailyspeculations.com/archieve/The%20Speculator%20-%20Why%20the%20trend%20is%20not%20your...

And more recent academic arguments in favor of T/A:

Excellent short overview:
http://intl.econ.cuhk.edu.hk/topic/index.php?did=11

"Most economic studies concerning the profitability of technical analysis find favorable results. For example, Fama and Blume (1966) studied the profitability of filter rules for the US stock market as a test of random walk or efficient market. More recently, Brock, Lakonishok and LeBaron (1992) investigated the profitability of simple moving average trading rules and trading range break-out; their results provide strong support for the technical strategies. Following Brock, Lakonishok and LeBaron (1992), numerous studies have investigated the profitability of trading rules for stock markets around the world. Examples include Hudson Dempsey and Keasey (1996) on the UK market; Bessembinder and Chan (1995) on the Asian markets; Ito (1999) on the markets of Japan, U.S., Canada, Indonesia, Mexico and Taiwan. All of these studies focus on stock market indices instead of individual stocks."

http://216.239.39.100/search?q=cache:LwP2Jdg-pfEJ:www.sta-uk.org/rg_wita.pdf+brock+lakonishok+lebaro...

"Brock.W ,Lakonishok.J and LeBaron.B (1992) `'Simple Trading Rules and theStochastic Properties of Stock Returns' Journal of Finance,47,1731-1764, confronted the efficient market hypothesis, and showed that returns obtained by simple yet popular trading strategies used in technical analysis were not consistent with the random walk model and the GARCH family of models upon which most of the testing of financial theory is based."

Additional academic research:
http://citeseer.nj.nec.com/context/516866/0

http://www.sta-uk.org/rg_wita.pdf
http://econ.ucsd.edu/papers/abstracts/9731.html
http://finance.wharton.upenn.edu/~kavajecz/tech020516rfs.pdf

The case against technical analysis:
http://www.supertraderalmanac.com/censorship/technical_analysis_deemed_fraud_.htm

According to the government, "[R]espected scholars are virtually unified in their recognition that even the most legitimate technical systems (with their hypothetical and retroactive foundations) are incapable of providing the trader with any significant market advantage."

"The efficient market capital model emphatically contests the notion that financial markets are so inefficient that speculators can exploit these markets' inability to adjust to all types of information. Although the limits of the efficient capital market model, and its implications for regulatory policy, are a dependable source for endless debate, few dispute the model's general predictive powers. In fact many important regulatory policies are predicated on the model's accuracy.

"Virtually the entire economic community is in agreement, however, that the efficiency of the market is sufficiently strong so that all publicly available information is rapidly disseminated and is then almost instantaneously reflected in the price for any widely traded investment contract. As a consequence, investor analysis of specific investment contracts will not lead to superior gains, since it will require an analyst to predict value better than the market as a whole. Thus, while some traders will profit while others will lose,
the outcome of speculative investment is unlikely to significantly outperform chance. See Dennis, Materiality And The Efficient Capital Market Model: A Recipe For The Total Mix, 25 Wm. & Mary L. Rev. 373 (1984); Posner, Economic Analysis of Law, Ch. 15 (4th ed. 1992); Comment, The Efficient Capital Market Hypothesis, Economic Theory and the Regulation of the Securities Industry, 29 Stan. L. Rev. 1031 (1977); Fischel, Use of Modern Finance Theory in Securities Fraud Cases Involving Actively Traded Securities, 38 Bus. Law. 1 (1982); Lorie & Hamilton, The Stock Market: Theories and Evidence (1973); Fama
(1970)."

"Technical analysts . . . first make a deterministic (one might say spiritual) leap of faith that non-random price patterns exist. They then illogically posit that these patterns, once revealed to the few (or indeed -- through marketing -- to the many), may be successfully exploited in trading. To accomplish this, of course, the 'pattern' must remain undetected by others (otherwise the increased market activity defeats the 'pattern' by driving the price to a point where speculation is no longer profitable). See Marshall (1989) at 263-264. Public policy presumes that markets are not so witless. 'The presumption is [] supported by common sense and probability [as] recent empirical studies have tended to confirm Congress' premise that the market price of shares traded on well-developed markets reflects all publicly available information . . . '

The author of this text:
http://www.supertraderalmanac.com/censorship/technical_analysis_deemed_fraud_.htm
is Frank Taucher (First Place Winner of "The $40 Million Investment Challenge" for having produced the highest verifiable investment return, 1980 - 1993, as reported in "Forbes" magazine and "The Wall Street Journal"):
http://www.supertraderalmanac.com/

Martin "Buzzy" Schwartz (“the best there is”):
http://www.annonline.com/interviews/980422/biography.html
"When he began trading stock options in 1979, "Buzzy" earned twice as much in his first year ($600,000) as he had earned over the previous 9 1/2 years as a security analyst.

Schwartz attained fame in the trading world through repeated entries in the U.S. Trading Championships run by Norm Zadeh of Stanford University. In these nine contests, Schwartz earned more money than all other participants combined, with an average return of 210 percent, nonannualized."

http://www.amazon.com/exec/obidos/tg/detail/-/0694519316/qid=1055639193/sr=1-9/ref=sr_1_9/002-823618....

http://www.amazon.com/exec/obidos/tg/detail/-/0887306101/ref=lib_rd_ss_TC01/002-8236185-7059218?v=gl...

http://www.jameslevine.com/Moreinfo.asp?idty=46

http://www.smr.com/pitbull.htm

Quotes by Marty Schwartz:

"The marketplace is an arena and other traders are the adversaries.

I turned from a loser to a winner when I was able to separate my ego needs from making money. When I was able to accept being wrong. Before that, admitting I was wrong was more upsetting than losing the money.

When I became a winner I went from 'I figured it out, therefore it can't be wrong' to 'I figured it out, but if I'm wrong, I'm getting the hell out, because I want to save my money and go on to the next trade.'

By living the philosophy that my winners are always in front of me, it is not so painful to take a loss. If I make a mistake, so what!

My attitude is: Never risk your family's security.

Whenever you get hit, you are very upset emotionally. Most traders try to make it back immediately; they try to play bigger. Whenever you try to get all your losses back at once, you are most often doomed to fail.

After a devastating loss, I always play very small and try to get black ink, black ink. It's not how much money I make, but just getting my rhythm and confidence back.

Before taking a position always know the amount you are willing to lose.

The most important thing is money management, money management, money management. Anybody who is successful will tell you the same thing.

I always take my losses quickly. That is probably the key to my success.

The best advice I can give to the ordinary guy trying to become a better trader is Learn to take losses. The most important thing in making money is not letting your losses get out of hand."

Marc, I am disappointed by KT's performance this year:
http://www.geocities.com/k_tieff/e/KT-QQQ-e.htm

After today, KT is up only 18% for 2003 and his newly modified system is more conservative than the one he used before.

Marc, exactly right...reacting to the market is just as important (if not more) as predicting it...SOX needs to reverse up for NDX to move up again...

Robbins World Cup Stock and Futures Trading Championships Winners:
http://robbinstrading.com/worldcup/stocks/standings.asp
(the best performers in any category - stocks, options, futures - exceeded 315% return in only 5 out 19 years)

Robbins 2003 World Cup Stock Trading Championship Agreement:
http://www.robbinstrading.com/pdf/wccst2003.pdf
(minimum deposit: $15,000, duration: calendar year 2003, securities: limited to stocks and options)


(My view: I believe that it may be worthwhile to enter this contest for 2004 with a good chance of finishing among the top 3 in the stock division if one can achieve 100-200% annual return...something to seriously consider as 2003 progresses provided the results for 2003 are superior)

Mark D. Cook, 1992 U.S. Investing Champion (Option Division):
http://www.markdcook.com/registered_only/articles.htm
http://shop.store.yahoo.com/traderscom/v19364ininch.html
http://www.traders.com/Documentation/FEEDbk_docs/Archive/032001/Abstracts_new/Interview/Interview.ht...

The 1992 U.S. Investing Championship included Mutual Fund Switching Category:
http://www.markdcook.com/articles/USInvestChampOptions.pdf

Bo Karag, 1993 U.S. Investing Champion (Option Writing Division):
http://www.investortradesonline.com/
http://www.investortradesonline.com/id5.html