I feel like my psychology is finally at the right level to really take my trading to the next level. Let me share what I mean…
I demonstrated phenomenal winning ability to predict markets– at level that I am not sure that I have seen anyone else demonstrate. Of course, much has never been published. On top of that, I built top ranked futures systems that really performed well.
Yet, my personal trading has never fully reflected my abilities.
There are a myriad of reasons that I have lost money over the years. The question is whether or not there is a root cause? I think there might be.
- I have always treated my predictions as very important. I was never as critical of my trades. This means I rarely make predictions unless I am very confident I can win. As such, they tend to be lower frequency. I win my predictions at very high rate. I do not hold my trades to the same standards: sometimes they are scalps, experiments, etc. I remember a vendor from back in the day, name momentarily eludes me, who had a saying, “I can spit a pot of gold.” This saying made me think about the nature of my predictions– there is power in having to make a call.
- My confidence is my predictions has been reliable over they years: however, my confidence in my trades has not been as reliable. There is a more critical aspect here too though– my ability to trade with heavy discretion is based on accurately assessing the odds. When I have traded more actively, I have not always maintained the ability to accurately assess my odds. If I were truly calling 85% trades, which is probably below my average prediction win ratio, then the probability of 2 losses in a row would be 2%.
- Correlated losses. Correlated losses have been one of my biggest fears because it has been a major cause of losses. Because I can normally call the market with high accuracy, sometimes the greatest risk is that I take too many losses pursuing a single idea. On the other hand, I do know that staying with a trade will often pay off. At this time, I am of the mindset that if a trade doesn’t work I should move on– 2 strikes at most.
- Discipline, situational awareness, cognitive bias, and tactical awareness. Discipline was rarely a problem for me and I felt I had mastered that aspect. However, I found problems resurfaced in specific circumstances: in other words, I did have discipline problems but they only occur in specific situations. When I traded high frequency or too many contracts, often trying to work a trade for huge payoff, I would sometimes lose track of my risk.
There are many causes of problems here. In one case, I had a cognitive bias. I was able to call a micro-scalp with accuracy and a trade to a target with high accuracy. On either type of call, at the time, I was probably 80% or better.
So, this gave me the idea that I can trade risk free– load a ton of contracts on my micro-scalp and ride it to my target trade. During this particular time, I took heavy losses as I got into winning scalp after scalp and all my target trades worked. How could it be? A very simple cognitive bias. Most small or micro moves mean revert. Think about how any market trades in a range. Most moves mean revert and do not jump to to the next level. In order to make this style work, it is not enough to call the micro movement– you would have to pick the exact micro movement that led to the jump. If I had just traded the scalps or the target trades, I would have been very profitable over that period.
Another type of combination of risk free and situational awareness came from counting my open profits against new adds. In this case, I am working a trade for a bigger profit. My goal is to put on and take off trades as the market moves toward the target. As I take add and take off, my break even can move upward towards the target. In very volatile situations, it led to my swinging from a large profit to large loss. This only happened once in extremely volatile situations. However, similar problems can occur if not tracking commissions accurately or if the trading software doesn’t count the p&l as you think it should.
Another type of risk free illusion comes from a specific trade where I attempt to size up to much larger then normal size. I only traded this way when I was sure that the market wouldn’t tick back against me. However, there is a fair chance of a “micro whip back”. As such, I still need to leave it room. The illusion here is that because most trades don’t whip back that there is no risk on the trade. Clearly, it is not risk free and the stop will be hit sometimes. But, this is a specific trade I got carried away with at times.
Another form of cognitive bias with the risk free trade illusion has happened to me when I knew I had a strong quantitative bias. In this case, I know the odds for how the trade will resolve with very high accuracy. And, if I traded the system as designed then those odds would be reflective of the reality. However, at times, in this situation I have tried to add more size. Adding more size would not change the expectancy if I kept all else equal: though it would cause me to risk too much. But, I would tighten the stop which would change the odds. This gets into tactics of when one can add size and how it should be added. There are many technical aspects here that I had to learn over time.
- Technical problems & lack of preparation. In one period of time, I was trading in an arrangement for another. As a result, I did not have control over the setup. There were certain technical problems that I brought attention to the backer. The backer, a good friend to this day, did not take my concerns seriously. It ended up causing several large losses right off the bat simply because of the software. This was a major and very professional “third party” software connected to Interactive Brokers.
I never had technical problems trading for a over year (at that time) with the defunct OEC. Well, I had 1 hung order over a period of a year of very active trades. While we eventually resolved the problems, the losses made things more difficult. And, more recently I have never had problems with Ninja and AMP, either.
I do know a professionals who use this software– which honestly is very similar to Tradestation and might be better in many ways. So, I am not sure if it was terrible luck or some combination of factors. What is important is that we could not test this stuff out in the sim because it was at the interface level between the software and the broker.
- Lack of capital & large futures notional value. This was a real problem. It has caused me to not to be able to trade my first years. And, yes it even caused losses when funded because I could only bring up the technical issue above with the backer but could not take decisive action like I would if trading my own capital. As well, some techniques did not translate well from large to small accounts. One of my core trading styles was based on developing a confidence in what was happening and then dynamically increasing leverage. This type of trading requires either a “small unit size” or a “large account size” and until the e-micros I couldn’t trade the same as I practiced in the simulator. Basically, trading a small account my smallest size was larger then my largest size on a larger account. Because I traded just like I trained when I did lose, I lost more then I should.
- Summit fever, optimistic thinking, assumption of skill transference. These are side effects of lack of capital. But also the inaccurate confidence assessment can come from thinking the skill transference from trading one way will translate to another.
- Poor position sizing. Trading strong ideas too small can be as bad as trading too big. Daily profit objectives can make sense but self limiting profits is often primarily psychological in nature.
- Lack of confidence. On sizing up quantitative trades, this gets back to a lack of confidence in my abilities because I would not feel the need to size up those trades if I had maintained confidence in my discretionary trades.
- Information loss. While my knowledge has been accumulative, I cannot trade the full extent of my knowledge without running many systems which take time to build & test.
- Lack of sustained focus. This is one of the reasons I have built the Quantitative Collaborative. One can either scale up or scale out– in either case it requires a lot of work.
- Unbalanced focus on trading performance. A great example of this was trying to trade the afternoon session after being profitable in the morning– knowing I do not do well in the afternoon– in order to get in “training” and improve my “performance”. The goal of trading must be to make money or it is not going to be sustainable. In fact, no reason at all to do this– given I could have traded on simulator.
So, we see that I have experienced losses for a myriad of reasons. However, if I had to break it down as to the most important causes then it has to be the following as most important:
- Inaccurately assessing the probability and quality of my trades. The simple fact is that if I were winning 85% of the time then my probability of 2 losses in a row would be 2%.
- Poor position sizing. The encompasses both too big and too small size. This also address correlated losses– as sizing down would limit that risk.
- Poor execution. This gets to a lot of the problems caused by sizing, working orders, or tactics.
- Inadequate specific preparation and plan with new techniques. This also gets into impedance mismatch. One should trade in their comfort zone– losses often happen when one gets outside their comfort zone or starts trading at the wrong level.
- Lack of proper context and focus on the root goal of trading: to make money. Too much enthusiasm for trading performance and short-term goals.
The real answer that must be taken into account is that trading, at an elite level, is not simple. Yes, a master can make it appear simple but that’s what masters do in any discipline.
One could state the root cause of most of my losses has been a combination of not accurately assessing my abilities and trying to trade beyond my abilities. That is a true statement. There is no way around it. However, it is not entirely satisfying because if I never pushed myself from the beginning then I would not have developed to the extent that I have.
I think this gets at a really deep and important matter too. The obtaining performance aspect must be a stage in trading (or trading a new method) that should be learned in the simulator (or with trivial risk). And then when one goes live while performance development continues to be important: risk preservation and capital growth must be the root consideration. At the trading live stage, one should already have proficiency in the method, which could simply be defined as ability to extract regular income.
And, this might be one insight into why so few make it trading. Most traders start with the goal to make money. They do not have the drive to develop the skills. A smaller subset do become enamored by the allure of mastering markets without realizing that performance development should only be a stage in any method.
While it may sound strange, there is a corollary in other professions. Computer science students will often hand-craft common algorithms in order to understand how they work. On the other hand, barring exclusive requirements hand-crafting common algorithms, instead of using common packages, in a commercial project would be a sure sign of amateur work– because it would increase the maintenance cost, needlessly increase the time spent and expense to the client, and the custom work is highly unlikely to be as robust and complete as well-developed packages.
What is the breakthrough? Yes winning trading is easy when winning but otherwise….
One must be have a robust plan, multiple layers of risk control (daily loss limit, trades, size, correlated losses, technical problems), a strong position sizing logic, ability to focus edge, ability to know when edge is waxing or waning, and the ability to learn and adapt over time.
Importantly, the psychological insight is that one must have a balance between performance focus and income focus.
One shift has been away from generalized market processing toward pursuing specific quantitative edge. I will discuss this in next post.