Uncertainty, Knowledge, and Discipline
I happen to currently host some of the top long term tracked futures systems at Collective2, and Iâve seen a lot of high flying vendors come, soar, and blow up. One vendor, whom Iâll not name, lost all of his systemâs prior profits on holding onto a single losing trade which apparently did bounce back shortly after he sold out. He wrote something to this affect to this subscribersâ:
I knew it would bounce back. But, I received so much criticism that I went ahead and sold them.
If he knew it would bounce back then there it is absolutely not credible that he would have dumped his contracts which meant ruining his system. If he had held onto the contracts it would have looked terrible and no serious trader would have subscribed. But, he might go onto to new equity highs and attract the less realistic and informed subscribers. Obviously, if we believe the statement then we are left with an individual who isnât rational, an individual who could have easily and surely made money who decided not to. I think the correct statement has to factor in uncertainty, and I imagine a more truthful statement would have been:
I thought the market should bounce back. However, I have already lost most of my systemâs profits waiting for it and it might go lower.
The real underlying issue was not the traderâs discipline but the uncertainty of not knowing what the market would do. If you canât embrace uncertainty then youâre always left second guessing yourself and focusing energies on matters that canât resolved. There is always going to be uncertainty in trading. I think that most traders try to eliminate the uncertainty which is always a losing proposition because with no uncertainty then there canât be any profit opportunity.
Developing the ability to make sound decisions under uncertainty is the critical skill of elite trader development. Most problems that deal with discipline arenât discipline problems so much as uncertainty problems or even one might think of it as discipline under uncertainty. In Gary Smithâs book âHow I Trade for a Livingâ he wrote about how one trader needed all of his indicators to line up before he placed a trade. Some of the best traders donât even use indicators. See an indicator is a crutch. An indicator is something that tries to take the uncertainty out of the market and if it worked then it would be great but they donât. They donât take the uncertainty out of placing a trade. So, the trader who thinks the problem is to remove the uncertainty will direct all of his energies into a futile task. The trader who instead works on making his best decisions even while uncertain will be well served. To work on this, reduce the number of indicators you use, reduce your size, simplify your stop usage, and then work on taking every trade idea you have and on making quick decisions about the market and acting on them. Trade very small, trade as often as worthwhile, and trade decisively.
Of course, you may agree that the problem with selling at the bottom was due to uncertainty. But, would not his oversized position be indicative of a disciplinary problem? Perhaps, yet I would want to consider if he had the knowledge to know that he was oversized. For example, I happen to overweight and that means that I eat more then I burn. Yet, if I knew precisely how many calories that I was taking in and how many calories that I was burning then I would always keep a neutral or negative total. In this way, the traderâs problem would only be a discipline problem if he knew the maximize size that he could trade without going over the edge. But, he if didnât know that then it could be a problem with knowledge or for a better term: the lack of knowledge. In his case, I do recall his position sizing was crazy stupid. But, even if one isnât taking crazy stupid actions there are a wide range of reasonable leverage amounts based on different estimates. For example, one of my mechanical trading strategies probably could have been traded the last 3 years with only 2-4k. Some years before then one would probably need 5-7k. The level of leverage one could use has varied considerably based on the market conditions.
One has to somehow develop knowledge. If I recorded precisely what I would eat over a day then I would start to have better knowledge, not a perfect knowledge, but a better knowledge. In the same, we can never have perfect knowledge in the market. But, we can develop a better knowledge by reviewing, comparing, and contrasting the current market to historical markets. One of my bread and butter trades is the ability to call 2 to 3 day market swings whether thrusts or reversals. This is a discretionary trade so I didnât have a good estimate on the type of risk involved in holding overnight. I created a âperfect traderâ in order to look at historical performance. Basically, I created a script to look into the future and pick out perfect trades according to my criteria (i.e. only wining trades, only 2 to 1 risk reward) and then I went back and made trader only win 60% or 70% of the time. In this way, I was able to get more knowledge about how that worked. I was surprised at the extent of the drawdowns of the perfect trader. The real purpose of this exercise was to gain some additional knowledge. Yes, a hypothetical and rather fuzzy knowledge but at least I could know something.
There are a lot of ways to gain knowledge about the market. Most people claim to be students of the market. But, they are really students of false prophets: that is students of books, snake oil gurus, movies, and media. Below I list some ways to help one develop knowledge:
Build the âperfect traderâ that trades your style (that is use a script to look into the future to pick out winning trades) use this perfect trader to generate useful ideas about maximum performance, risk, and other metrics. Review your indicators and see how often the perfect trader uses your indicators and how he uses them.
Build a fully 100% mechanical trading strategy. Use the performance metrics to gain an estimate of future probability. Warning: It takes extensive experience to build working mechanical strategies.
Measure the average range of your market over various time periods.
Check to see how your market is correlated to other markets.
Review professional trading disclosures for ideas and to see how others using similar strategies have performed.
Use options pricing or random pricing models to get an estimate of the probability of a stop hit.
Find the factors that influence your performance and start to measure them. Some ideas: measure volatility, the number of large trend days over recent x period or versus a historical average, numbers of days without a significant pullback, etc.
