I received a great question from Jason:
Mr Grimes, I need to ask you something. I have been watching your videos and I’m a little confused about one thing. On one of your examples, you showed a diagram of a collection of people making I believe 250 trades with a wining process. The diagram showed that a certain amount of people still lost money and you made a comment that winning and losing was random, even with a successful process. This information really made me take a week off to think “Do I really want to do this with my life?”. I know the market is unpredictable but even with a process, risk management, and slight odds in my favor — I could still lose all my money even after 5 years of trading small like you recommend? Are you saying that winning and making money is completely random? Thank you for clarifying this.
This is a great question that every trader will grapple with, more than once, at some point in his career. I hope my answers in the podcast are entertaining and useful!
I also want to link to a few resources that will help you dig deeper.
Michael Mauboussin’s book The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing
I also wrote a two blogs, here and here, that will give you a slightly different perspective. In particular, the second post has some ideas for managing your reactions to luck and for accepting the impact of luck in a constructive way into your trading process.
If you enjoy the podcast, one of the very best things you can do for me is to leave me a review on iTunes here. Also, if you like the music for this podcast, then be sure to check out Brian Ashley Jones, my friend, and a fantastic singer-songwriter.
Enjoy the show:
The conclusions of that study sounds like a contradiction. If all the 250 people were truely using the same winning process then they should all have the same results because it would be equivalent to 1 person with a winning process and exploding that data set 250x. I think what this study really shows is the difficultly in perfectly emulating another persons trading strategy and the kinds of intangibles inherent in trading systems.
You’re getting hung up where many people do: on the assumption that a trader with a winning process will make money. That is precisely the point of the article. It’s a hard lesson, but it sounds like one you’re avoiding. 🙂
For the record, the example was simply a random draw with fixed win/losses, as I explained in the podcast. What this example illustrates is how much variability there can be, even in a “winning” system.
Thanks for clarifying the premises Adam, unfortunately I did not have time to listen to the podcast. In this case I do agree with the conclusion and it is not a contradiction. I had assumed that this was a pseudo-random initial condition, something like 250 traders trading the same stocks simultaneously with very similar systems. What you’ve described now is like a monte carlo simulation with 250 individual walks. The outcomes here will certainly depend on the n-size, w/l amounts, and w/l probabilities, and you should get a probability distribution.
Thank you for recommending the “Success Equation” book. I was dumbfound when I read the “Peak” book by Anders Ericsson and he couldn’t interpret his own research correctly (“Successful people work hard? Only hard work matters!”).
A thought, correct me if I’m wrong: It’s possible to essentially get rid of this problem if you are day trading AND diversifying – instead of risking 1% or 2% in 1 instrument, you risk 0.2%-0.4% in 5 different instruments (say, ES, GC, CL, 6E and 6J) and since their intraday moves are not very correlated (from what I’ve seen, although I don’t know how true that is in crashes yet), you should rarely have a losing week, let alone a losing year (provided that you make, say, 3 trades a day per instrument, which is 300 trades per month).
That’s what I’ll be looking to do once I make sure my edge is robust. Also, maybe a topic for a future podcast: can you talk everything about edges and yearly returns? What is a realistic discretionary edge and yearly returns (IIRC, you spoke about how 10% edge is realistic but that it’s possible to get quite a bit higher than that), and how do those figures chance once you start trading on the lower timeframes?
I ask you this because I had tested my discretionary system for around 50 trades, got something ridiculous like a 70% edge but, after a few more hundred trades, and once I eliminate the 1 day that was clearly a positive outlier, it looks like my edge is in the ballpark of 5%-6%, with 3 trades per day. If I always risked 2%, which is too much for day trading (even on a demo account, can’t stomach drawdowns of 12% that happen in a few hours but diversification would solve this to a certain extent), it would be something like a 75% returns/year without compounding. Given that, should I expect my edge to become smaller?