Volatility and the Trader

I was told when I first began trading that ‘volatility is the life blood of a speculator’. Like all trading truisms, this has an element of truth. Imagine trying to eek out a living when ranges are tiny and market direction flat.

But there are times when markets become ‘too volatile’: the meaning of ‘too volatile’ is personal and varies from individual to individual. The important thing is to have some measure of ‘too volatile’. My definition is two consecutive days of ranges greater than an ATR of mean + 3 standard deviations. The definition is instrument specific so that I can have, say the ES being too volatile while BPJY may not be.

Once I see a ‘too volatile’ reading in an instrument, I cut all positions (usually profitable ones; losing trades have probably been stopped out). The reason I do this is twofold:

  1. Quantitatively, my setups are validated by price action that is ‘normal’. When markets get too volatile, the population has been too small to date to draw meaningful conclusions. So in accordance with my first philosophical rule for trading (preservation of capital), I take myself out of the market.
  2. Qualitatively, my psyche is comfortable with only so much volatility. Beyond a certain point, I know I am likely to think the dollar value of the tic fluctuation rather than dwell on the info the market is providing. I can become accustomed to increasing volatility if the increase is gradual. But too rapid an increase places my in a zone of discomfort.

The message I am pushing here is take care if you are feeling uncomfortable with the ranges and market movement.

In line with the theme of this post, let’s turn to the ES.

To identify the boundaries for this structure (in Market Profile Terms – distribution) are off the 13-w swing low prior to the sideways price action (development). I notice that although the swing low occurred June 14 2006, the directional move did not start until July 18 2006. I prefer to anchor the July low to start my retracement levels.

The Primary Buy Zone lies within 1272 to 1229 (basis the CSI perpetual contract). The current difference between the CSI-p and ESH8 is about 5 points (ES is lower than CSI-p by 5). This makes the Primary Buy Zone 1267 to 1225.

01-22-2008-sp-18d.jpg

FIGURE 1 S&P Primary Buy Zone

Last night we edged into the Primary Buy Zone. This is an area where I usually par my position size to no more than 1/3 the initial size.

In addition, the last time the market reacted so strongly, Greenspan came in and cut rates prompting a 5% rally. Now this does not mean Bernanke will cut rates today but he might. Whether or not he does, will only provide more fuel to a market that is volatile.

All in all this is a good time to put some $ in your pocket. Oh sure, the market may head lower but then again it may not. I’ll part with this story: A student was long gold and sitting on some very nice profits. He was looking to exit at the Primary Sell Zone. The market got to within about a dollar of the minimum price.

To say he was excited …well that would be putting it mildly.  As he pointed out, the weekly bar looked good (opening near the lows and closing near the highs). Based on it, there was every reason to believe that the profit levels would be reached. However I did warn him that the daily’s painted a weaker picture and that he should establish some level beyond which he would cut the position.

He decided to place a breakeven stop but I could tell that he did not feel his stop would get hit. Well it did. The market opened lower on the Monday and proceeded South.

The point of the story is this. If you have profits in the ES, Gold etc, identify a level beyond which you will cut the position: Nothing feels worse than letting large profits turn into losses. When I say ‘identify’, I mean treat the loss as having occurred – feel it – rather than just give lip service to the level.

S&P Blog Monday Jan 21

Sorry for the break in continuity. I guess I am too unfit to be travelling as much as I do. I’m either going to have to get fitter or stop travelling as much. For the moment, I have opted for a personal trainer to create the habits to improved fitness!

Today I shall be writing about the S&P but whereas in the past, I focused on my thought processes, today, I’ll focus on my tools. Figure 1 shows the matrix of tools.

prduen-matrix.jpg

FIGURE 1 Matrix

I replaced the matrix with this one from Hank Pruden’s “The Three Skills of Top Trading”. The difference between the approach I use and that of Hank’s is Hank uses a more quantitative process. The matrix serves as a check list to make sure I am trading what I see rather than what I hope to see.

In the S&P, the most sensitive of my indicators, the Sentiment Indices are beginning to register oversold readings. We need to remember that these tend to lead prices by at least 2 weeks and up to 6 weeks. I use the tools to put me on notice not to get too aggressive. The time to be aggressive is when the Sentiment Index is no worse than neutral.

My main tools are in the Price Matrix. The yearly trend (12-month swing), has registered a possible Upthrust Change in Trend – an uptrend that began in 1982. However the Volume configuration is not there: the high on the Oct 2001 high was much higher than the volume on the Mar 2000 high. Nevertheless, if the signal is confirmed by a monthly close below the Primary Sell Zone (1455), we can expect minimum target of 787 to 613. Figure 2 shows the 12-M swing

12m.jpg

Figure 2 12-M S&P

The 13-week (quarterly trend) and the 18-day (monthly trend) need Whole Point Counts (WPCs) at or below their respective break down point to confirm the change in trend in those timeframes (see Figures 3 & 4). Strictly speaking we don’t need the WPCs given that the 12-m line has turned down; but I prefer to see WPCs in the respective lower timeframes as a final confirmation.

13w1.jpg

Figure 3 13-w S&P

18-d.jpg

Figure 4 18-d S&P

If the trend is down, what would invalidate the possible signal? Preferably without the WPCs forming, acceptance above the Primary Buy Zones of the 13-w and 18-d. These are 1431.6 and 1422 respectively basis cash.

Volume and momentum all suggest further downside. The Ray Wave structure suggests 1289 to 1290 (basis ESH8) a possible area for a bounce. The 5-d impulse stats show the 5-d line has gone almost mean + 3 standard deviations as at Fri’s close. The cycle services I subscribe to indicate a cycle low this week, Tues to Thurs. If we do get a bounce, the quality of the bounce will tell us whether or not a bear market has started.

Assuming the bounce has a bearish structure, a preliminary resistance area would be 1366.5 to 1354.5 (basis ESH8). The next level to watch is the 1388 (basis ESH8), the start of the directional move down on the 30-minute Market Profile (See Figure 5). The 1388 is well below the Primary Buy Zone but given the momentum and volume to this downside move, acceptance above the 1388 would give me cause to re-evaluate.

mkt-profile.jpg

Figure 5 30-Minute Market Profile ESH8

The question I am usually asked is: “I did not get short before the break. Should I short the market now?”

My answer: “I don’t know what the market will do. But on a risk-reward basis, with possible support coming in for both price (statistically for the 5-d and on the Ray Wave targets) and time (cycles, this week), I’d treat a short now as a high risk rather than low risk trade.

The time to get short was when the market failed at the top of the Value Area providing a possible Ray Wave Failed 5th (See chart 6). For now, wait for the bounce, whenever it may come but it will come.

18d-failed-5th.jpg

Figure 6 Ray Wave 18-d Failed 5th

A Review of the S&P 01-09-2008 II

I often get asked: “Why do you set out so many scenarios when preparing your trade? Why not create a simple plan e.g. buy at xxxx.xx with a stop at xxxx.xx?”

I answer: “It’s because I don’t know what future will bring”.

By this I mean how a market does something, is at least as equally important as how it does it. Yesterday’ price action in the S&P is a great example of what I mean.

In yesterday’s blog, I set out four scenarios. The one I felt to be most probable was the one speculating on a smaller than normal range day. At time that I was writing the piece, the market was up about 9 points. Hence I focused on what I would look for if the market broke the highs. I suggested that the level representing 50% of the gap would hold. This was the second favoured scenario.

However, by the time the market opened yesterday, it was only 1.5 points up. At the open this is what we knew:

  • Tuesday’s high was 1436.5 and its low was 1396.5;
  • On Jan 9, Wed, the ESH8 opened at 1398.

Given the above, at the open on Jan 9, the greatest probability occurrence was the 1st scenario that provided for a small range day but now we’d be looking for a breach of the lows. Notice that although, I shifted from looking for a breach of the highs to one looking for a breach of the lows, the essential nature of the scenario remained unchanged – I was looking for a smaller than normal range day.

What does ‘normal’ mean? I define normal as being mean +1 to mean 0.5 standard deviation.

A more important point: recall that I had, as a third scenario, canvassed the possibility that the market would break to new lows. For the reasons stated in Tuesday’s blog, I had decided not to participate in any further shorting.

Now, let’s turn to the picture when yesterday, the market broke to new lows on two occasions: (Market Profile’s) ‘E’ period and then the ‘I’ period. When we broke for the first time, (‘E’ period), we had a True Range of 16 points. This was in keeping with what I was expecting – a smaller than normal range. Mean 25 points and we have a standard deviation of 10. Hence I was looking for a range no more than mean -0.5 stdev i.e. 20.The ‘I’ period extended the True Range to 22: we now had normal and when it rotated back into the day’s range, it raised alarm bells for my short positions.

Here’s what I noted:

  1. Yesterday was the 9th consecutive day of lower lows and lower highs. On a 1-day swing, the market was statistically overbought.
  2. The market was in the Primary Sell Zone: an area that would support a rejection of the down move IF the uptrend was still intact.
  3. If the market now extended the range up, we’d have a Market Profile Neutral Day. If a Neutral Day closes in the middle, we have a balanced day that warns us of a possible trend change; a close in the top quartile is stronger evidence of a reversal since it provides for Free Exposure.

So when the market returned to the day’s range, I decided to:

  • Place stops on half my remaining shorts above the day’s high
  • Place stops above the Tuesday’s high on the balance
  • Exit if the market closed in the top quartile.
  • Go ¼ size long if the market closed in the top quartile.

Given the close, I ended the day being ¼ size long. I’ll manage the trade using Free Exposure guidelines.

The blog today I feel is one of my most important ones. It sets out how I manage trades using present tense information.

Chart 1 shows the Profile

blog-01-10-2008.jpg

Chart 1 Market Profile

A Review of the S&P 01-09-2008

I believe we have tipped over the edge on the S&P. In this post, I’ll set out why I say this and suggest some parameters to monitor.

To summarise this blog, the Ray Wave count suggests that we have topped out with a 5th Failure and ‘normal’ technical analysis’ supports the view.  The one drawback: The Whisper Number’s sentiment reading is oversold as are the other SI readings but not to the same extent as Whisper Numbers; this is an amber signal for me not to get too aggressively short.

ANALYSIS

The chart below shows the Ray Wave Count on the quarterly trend (13-week swing) 

sp-rw-13w.jpg

CHART 1 13-W Ray Wave

In the next chart I zoom in on the price action and go down to the 18-day (monthly trend)

sp-rw-18d.jpg

Chart 2 18-D

The S&P reached the Primary Buy Zone at 1370.6 to 1431.6 (basis cash) after reaching a swing on 12-11-2007. But note that the swing high failed to reach the minimum 78.6% retracement area in an established sideways market.

Ideas corner: when you are in a sideways market, the market moves from high to low, low to high; if the market returns to an extreme without first achieving the opposite extreme’s minimum target, we can expect the opposing the price boundary to be breached. In this case, we are at the Primary Buy Zone without having reached the minimum upside target. I am looking for 1370.6 to be breached.

Chart 3 shows the ESH8. You’ll notice that the price action on Jan 4th created a gap on the day the market popped into the Primary Buy Zone. The question in my mind was whether the gap was common gap or a breakaway gap. Last night’s price action suggests the latter.

When yesterday, the market open gapped up after a rest day, and failed to close the gap in the 1st 90 minutes, I thought the S&P would rally. I closed out my shorts and went long at the bottom of the developing value area with a stop for the longs below the lows of the day. You know I got stopped out. However once the lows were violated, the probability of the gap being a breakaway gap substantially increased. I waited for a rally to reinstate and increase my shorts. 

sp-es.jpg

Chart 3 ESH8

SCENARIOS

1) The S&P has been down for 8 consecutive days (excluding one inside day); it’s time for a breather and I would expect another inside day today. For this reason, I have created Primary Zones base on yesterday’s open and close. I’d expect those zones to hold.

2) If the market breaks above yesterday’s high, then the key areas to watch are 1447 to 1443 (basis cash). Apart from being 50% of the gap, there are a number of other ratios coming into this area.

Acceptance above 1147 suggests the gap will close. That in turn suggests the market is heading back up the Primary Sell Zone. I rate this the lowest probability scenario.

3)  A more likely scenario is the market will break above yesterday’s high and the 1447 to 1443 zone will hold. Should the market reach 1447 to 1443, I’d lean against this area to go short and look for intraday setups and triggers to enter.

4) The second lowest probability scenario is another strong day down. Acceptance below yesterday’s low on volume would warn us this is occurring. Since I dislike jumping on board a directional move once the 1-d swing is statistically overbought, I’ll give trading the ES a miss if this occurs; I’ll be content to manage my current short position.

Best of luck and do take care!

Routine & Habits VII: The Routines and Habits

Well folks this is the biggie. All I have written before in this series comes down to this post.

It’s my belief that participants attending seminars would obtain greater benefit from a seminar if they adopted, for 30 days after the seminar, a set routine to internalize the seminar concepts. At the end of the 30 days, the routine would become a habit and the seminar participant is then free to choose whether to adopt all, some or none of the content. The key point is that until he has internalized the seminar content, he is not in a position to choose.

I remember my first Market Profile seminar. I flew from Sydney with two mates: one was a technical trader working as a technical analyst for a broking firm (let’s call him John); the other was a day-trader (let’s call him Paul).  Paul and I ‘lost’ John pretty early in the piece. By the middle of the first day, John was doodling and post session discussions revealed that he had taken the view that the Profile was not for him – it was too far removed from what he knew.

Paul and I liked the idea of the Profile but we struggled with the application. It took me 9 months before I got comfortable trading the Profile way – what enabled me to persevere was the fact that I set a daily routine to follow. Each day I set the goal for that day and followed the set routine.  Part of the routine involved reviewing whether I achieved the day’s goal and if not what I had to do the next day to complete it. I would not move to the next topic until I was satisfied I had mastered each day’s goal.

Trading routines have a similar objective. Their outcome is to produce habits of success. The routine you set is one that is personal to you. Below is the one I follow – it suits my personality; treat it simply as an example.  Experiment with a routine, content, time of day, order of activities until you find one that sits comfortably.

Monthly Routines: 

  • Review summary of the psychology and equity journals. Are there any empowering patterns? Are there any disempowering ones? I look for the patterns under the headings: setups, instruments, and personal behavioural patterns. For example: is there an instrument that bore the preponderance of losses for the month? If so, did the losses occur for a particular setup? What was the difference between the current environment and the most recent environment that I made money in this instrument? In this setup?
  • Review equity journals to determine if I am on track. I look to make about 22% per year so I am looking to make about 2% per month. If I am not on track, where can I make improvements? What has to happen for me to return to budget? Etc……

It’s important to understand that I make these enquiries from a stance of curiosity – there is no question of blaming myself for errors or feeling anxiety because I am behind the eight-ball. True if I have had 4 consecutive losing months or more than 12% loss in any rolling period, I take an enforced holiday of at least a week. But this is in the form of a breather to put a space between the losses and me and not as a form of self-punishment.

WEEKLY ROUTINE:

  • Analysis of markets with a view to preparing a short list of possible trades in the coming weeks and a watch list where trades are unlikely but possible if certain events occur.
  • Analysis of psychology journal – summary of daily entries for possible patterns (see above entry in Monthly Routines).
  • Analysis of equity journal – to ensure that there are no out of boundary losses.

DAILY ROUTINES:

  • Download and update data
  • Update psychological journal
  • Update equity journal
  • Prepare for trades – plan, and visualize
  • Ensure day’s activities are planned around trading activities.
  • During the trading day, a series of routines to ensure my ‘rat brain’ is not running my trading.

That’s it. I hope you have enjoyed the series.

Routine and Habits VI: The Pyschological Plan

What is Winning Psychology? For me it’s a combination of traits and tools that provide the environment whereby a trader consistently executes his trading plan. In this blog, I’ll look at some of the tools at the trader’s disposal.

The first is preparation: whether you are a day trader, or longer time-frame trader, preparation is essential. The preparation takes place firstly at the conscious level and then at the subconscious. I set out some of steps of the preparation in my previous blogs in this series. Once we have completed the preparation consciously, I recommend we visualize in the alpha state the various step of the preparation. It’s beyond the scope of this blog to describe the visualization process but there are any number of good books on the subject. By visualizing our planned response to the various scenarios, we reduce the probability that we’ll breach our plan.

The second tool I like is a bio-feed device. The one I use is the 3100 WristOx from www.bio-medical.com but there are less expensive devices that do as good a job. The bio-feed back device warns me when my ‘rat-brain’ is starting to take control and before it totally takes over.

The final tool is the trading psychological journal. The purpose of my journal is to identify empowering as well as disempowering patterns. For example: a pattern that identifies when we are likely to over-trade. Once I find the pattern, I like to visualize:

  1. The events that created the environment
  2. Visualize an alternative response for a number of days
  3. Check to see if the next real-time event results in new behaviour. If it does, no further action is necessary. If it doesn’t, re-do steps (1) and (2).

For empowering patterns, I seek to create the pattern each time I trade.

I like to create weekly, monthly and 3-monthly summaries of the various patterns. In this way. I find it is easier to spot recurring patterns.

Routine and Habits V: The Money Management Plan

The money management plan balances the risk of ruin with maximization of profitability. In other words, it seeks to give us the biggest bang for our investment dollar with the greatest measure of safety: the name of the game is to survive a series of consecutive losses.

The Money Management plan seeks to answer:

  • What risk shall I take on this trade?
  • Given my stop loss, and my risk assessment of the trade, what is the maximum size I can place? If you have different levels of size, then the questions are: What is the maximum normal size and what size shall I have for this trade?
  • If you trade more than one instrument,  what is the maximum portfolio risk?
  • What maximum loss will I incur before I take a rest from trading?
  • At what point do I make my profits available to my position sizing?

To answer the questions, you have two components to consider:

  1. A psychological component: the dollar amount we risk needs to be within our risk profile’s comfort level; otherwise the probability is we’ll not follow our trading plan. If your current level of risk is above your comfort zone, increase your size incrementally – slowly boil the frog technique. Raise the size so slowly that your increased size doesn’t cause your subconscious to send out distress signals.
  2. A technical component: the inputs to your money management algorithm. There is your trade results: your win/loss rate, your Avg$win/Avg$loss, your avg$win for longs, your avg$win for shorts, the mean and standard deviation of possible consecutive losses, your maximum drawdown, the means and standard deviation of possible consecutive wins, your high water mark, your maximum adverse excursion, your maximum favourable excursion. There is also the market volatility: I use ATR to measure volatility.

All these factors impact the amount of risk you take. If you want a ‘quick and dirty algorithm’, there is the Turtle formulation that considers only the volatility of the market and the amount you want to risk – however you decide that:

(% Capital to risk x Capital)/(S value of ATR) = # of contracts. For example, let’s say you have US$20k and you want to risk 2% and trade the ES. Let’s take a 45 day – the ES has an ATR of about 25. The $ value is 25 x 50 = $1250. So the number of contracts you can trade is:

(2% x 20,000) = 400/1250 = 0

That’s right, a US$20k is not enough to trade 1 contract in the ES. The formula can be applied to any timeframe.

Money Management is one of those subjects that can be as simple or as complicated as you want to make it. I recommend you start with the Turtle formulation and move on from there.

Routine and Habits IV (B): The Trading Plan

Today I am going to write about aspects of a discretionary plan. Discretionary plan are first and foremost a reflection of our beliefs about the nature and structure of markets. Someone who believes in the market efficiency theory will use a different plan to someone who believes they are chaotic.

I believe that markets are chaotic. Consequently, I believe:

  • Markets have a discernable structure
  • They rhyme rather than repeat i.e. there are patterns we can exploit but these patterns are repeated’ similarly rather than exactly’.
  • The market rhymes because the patterns are a reflection of the emotional tug-of-war between the buyer and the seller.
  • The context in which the patterns occur are critical to the plan.

A discretionary plan is also a reflection of our psychology including our appetite for risk; this refection is articulated in our trading philosophy. In Trader Vic–Methods of a Wall Street Master by Victor Sperandeo, I found a statement that mirrored my values:

  1. Preservation of Capital – this is the overriding principle
  2. Consistency of Returns – this goes hand-in-hand with (1)
  3. Superior Returns – only when (1) and (2) have been secured

My plan and results reflect the three characteristics; for example I use the Rule of 3 not because it increases my bottom line. Indeed, in a strongly trending market, the rule reduces my profits. I use the rule because it smooths my equity curve.

I believe a discretionary plan has certain critical elements:

  • A way of identifying the trend of a timeframe and when the trend changes or is likely to change. Once we identify that a trend is likely to continue, or that it is likely to change, we have our strategy. That strategy is rooted in the timeframe we are trading and includes the effects on our the trend by higher time frames.
  • Tools to identify price levels where a trade may take place (zones) . As my nature favours a responsive trade, I look to buy support in an uptrend and sell resistance in a downtrend. I very seldom buy/sell breakouts of my timeframe.
  • Chart patterns that indicate a zone has held (setups) and entry patterns that tell me ‘now is the time to take a trade’ (triggers). The setup and trigger define the price stop placement. In addition to the price stop, I look to define the qualitative conditions that will cause me to exit a trade.
  • Chart patterns that define the core profit target in the Rule of 3. This core profit target versus the stop defines the Risk:Reward expectancy: I need to see around 2:1 or better to take a trade.
  • The relationship between the price stop, my risk assessment of the trade, and money management plan govern my position size. I have three sizes: normal, 1/2 above normal and above normal (usually 1.5 or 2.0 times normal).
  • A set of rules (Rule of 3) that govern my subsequent trade management i.e. the management of the trade once it starts to move in my favour.

What tools do I use?

  • TREND: Barros Swings and the Ray Wave
  • ZONES: Statistics of waves, MIDAS (see www.tradingsuccess.com free section’ for Paul Levine’s lectures on this tool), and various ratios.
  • SETUPS: Negative Development and Contraction
  • TRIGGERS: Intra-day volume on Market Delta software
  • INITIAL QUALITATIVE EXIT STRATEGIES: Based on Market Profile and Wyckoff
  • INITIAL PRICE STOPS: Barros Swings and The Ray Wave

Once you have your tools, you need to create a plan. I have found that classifying the rules under ‘Buy & Sell’ and giving a setup and trigger a separate rule number, is the best way of creating a data base to assess the efficacy of the rule. I’ll discuss this farther in the blog on ‘Stats to Keep’.

A couple of final comments. I believe all traders should have a passing acquaintance with statistical and probability theory. This comes from someone who was mathematically a complete dunce until well into his mid-30s. “Salvation” was found in two books by Derek Rowntree:

  1. Statistics Without Tears
  2. Probability Without Tears

Finally if you see yourself as a serious trader (as against someone having a ‘flutter’ i.e. a gamble for pleasure), you owe to yourself to back test your setups and triggers. Generally because a discretionary trader relies so much on context, it is difficult for most back testing programs to test the trading rules. However, you can certainly test the setups and triggers. In back testing, you are looking for evidence of robustness and if you lack the skills to do it yourself (like me), then find someone who’ll do it for you.

Routines & Habits IV: The Trading Plan

Plans can be fundamentally and/or technically based and if the latter, mechanical and/or discretionary. The fundamental based plans that I know of are all discretionary approaches; I have never seen a mechanical, fundamental plan but that doesn’t mean they don’t exist.

The best place to start is to define my terms:

  • Fundamental plans are based on inputs such as supply and demand, value etc
  • Technical plans are based on technical analysis
  • Mechanical plans: rule based plans, no trades are taken outside the rules
  • Discretionary plans: rule based plans with a rule that says ‘entries and exits need not follow the rules’. In essence, this allows intuition to play a role.
  • Subjective plans: a trading style that is totally based on intuition. Most of the pit traders I have met and some of their replacements, the ‘on screen scalper’.

I am a monthly trend discretionary trader – that’s my niche. But in today’s post, I’ll be writing about mechanical plans. I stress that in this area, my knowledge is vicariously derived – from books, my students and my peers who have been kind enough to share.

Before I get into the discussion, there are two points I want to make.

  1. I subscribe to a view put forward by Mark Douglas: that whatever style best suits an individual, the optimum path is first start as a mechanical trader. The mechanical approach teaches us to trade what we see, not what we’d like to see.
  2. The style we ultimately adopt is the one that suits our personality. In his latest video, Stephen Pierce makes the point that to succeed in business we first must choose the environment that provides the greatest opportunity for success (http://www.dtalpha.com/talkback/?p=17). Part of the trader’s environment is his personality. As we’ll see later, it’s not only in the realm of trading plans that a trader needs to take his personality into account.

Points (1) and (2) may seem contradictory, but my experience as a mentor suggests otherwise. Traders that move straight into discretionary trading more will confuse ‘intuition’ with ‘into wishing’. It’s worthwhile remembering that intuition is borne from experience of lessons learnt from our successes and failures. So unless you have built up the experience base, your ‘intuition’ is likely to be flawed.

The best mechanical plans I have seen have the following characteristics:

  • They contain around three buy and three sell rules: entry, stop and profit taking.
  • They are based on some observation about the nature of the market rather than  being based on just ‘data-mining’ i.e. some computer generated relationship.
  • Sound testing of the system for robustness is a must.
  • If the system trades more than one instrument, the testing needs to incorporate testing on portfolio basis. The testing program recommended was ‘Trading Recipes’; this has been replaced by Mechanica Software (http://www.mechanicasoftware.com/).

The best books I have read on the subject are the two by Thomas Stridsman: “Trading Systems and Money Management” and Trading Systems that Work”. Thomas makes a point that I have incorporated into the testing of systems. The testing for robustness is found in the normalization of results and not just by the dollars made or lost. He argues that $100.00 made on the S&P is very different from $100.00 made in oats because of the different volatility between the two instruments. We can take this a step farther and argue that the same can be said about the same instrument in different times. For example in 1987 a drop of 100 points in the Dow was a cause for concern; today, it happens almost routinely.

To normalize results, Thomas suggests we use on a one-contract basis, the result divided by the price initiating the trade: we have a result in percentages rather than dollars. In this way, we compare apples with apples between different instruments and different times.

By their very nature, the best mechanical plans ignore the context in which a trade takes place. Consequently, as long as the environment remains stable i.e. the system is operating within the conditions that suit it, it will make money. Indeed, given human nature, and taken as a whole, I believe mechanical traders will make more money than discretionary traders in this environment.

But in an unstable environment or one in a transitional phase, the mechanical trader will fare worse than the discretionary trader.

In the next blog, I’ll commence the series on discretionary trading plans.