From the Trading Floor to the Air: Lessons in Discipline
Mental Game of Trading
Featuring Dave Keller
Published on: August 6th, 2019 • Duration: 26 minutesDave Keller, CMT, president of Sierra Alpha Research, takes a multi-disciplinary approach to coaching financial advisors and institutional investors. In this compilation, Keller draws on his years of experience in markets as well as the lessons he has learned as amateur pilot to help others improve mindfulness and discipline in their investment process. Please note that this video is comprised of shorter clips that previously appeared on Real Vision, compiled here for ease of viewing.
DAVE KELLER: Hi there. My name is Dave Keller. I'm the president and chief strategist at Sierra Alpha Research based in Cleveland, Ohio. So, I help financial advisors and institutional investors to make better decisions. And do that in a couple different ways, help them maximize returns, manage risk, and bring more mindfulness and awareness to their investment process. I also enjoy incorporating non-financial topics into financial topics. So, I'm a student pilot, I'm a musician. And so, a lot of the best ideas I think you can bring as an investor are learning from other disciplines, learning from other activities. And I write a blog at marketmisbehavior.com.
So, I'm a student pilot. I've got about 80-some hours in a Cessna 172R, which is a pretty classic training airplane. And from the moment I started learning to fly, I would always chat with my flight instructor about the similarities between investing and flying. Because flying requires discipline, it requires taking the emotion out of it. You're often in very stressful situations, and you have to be able to function in that environment. So, the similarities to trading are hilariously common.
So, one of the ways my flight instructor always used to coach me was to tell me to be ready. And what that means is, anytime you were flying an airplane, you always had to have an emergency plan in your mind right in place. So, a lot of the flying process is done when you're still on the ground. And it's figuring out where you're going to go, registering a flight plan. And part of that is all along the way.
If I'm flying from point A to point B, where are all the opportunities where I could have an exit plan? Were there airports that I might go to? So, I might look for an abandoned airstrip, which in an emergency, I could probably get there. Or as you're flying around, you're always looking around, there's a golf course, there's a highway with no power lines. You're always thinking if something would happen right now, what do you do?
And the worst time that I- the worst experience of my learning that was when I first was starting to learn about emergency preparedness, your flight instructor will often just yank the throttle out. So, you're flying, everything's fine. 6000 feet, no problem. All of a sudden, the whole plane just goes. And it just drops very quickly. And he says, all right, you just lost your engine, what do you do? And of course, the first thing you do is you completely panic, you lose it and your heart starts beating, jumps up here, there's this whole physical sensation of falling, which is uncomfortable, and you feel like the plane's not doing what you expect it to.
And so, the first thing you do, the things jumping around there, too windy and flopping around them, trying to find some paper checklist. So, I can go through. We just lost the engine, what do I do, and I'm just slowly trying to pick all these things together, and I was a total mess at it. And again, it was totally safe. The instructor was there to make sure it's fine. But it was one of the most uncomfortable things I've done.
And then three months later he does the same trick, he pulls the throttle out and goes, all right, you just lost your engine, what do you do? And of course, at this point after doing a bunch, and I'm like, oh, I get that, that, that, that, that, that I need. You just know what to do. It's not a physical reaction anymore. You just know, yep, I lost my engine, this is what I do. I check this and this and this, and this and this.
And so, what you learn is in a stressful situation, and literally a physically stressful situation, which, as a trader, when you experienced loss for the first time is a very similar set of physical reactions. It's your heart beating more, it's uncomfortable, you start sweating. But you still have to make an emotionless decision or make a disciplined decision as a result. What do you do?
I always coach my clients to have a good exit strategy. When you put a position on, when you have a game planned in mind, what set of scenarios would cause that to be wrong? What market movement? Or what movement in that specific position would tell you that that is incorrect? And then what will you do about it?
And the reason why you do that ahead of time is because if you have a position and it drops 8%. Let's say you have a trailing stop, or you have a percent stop or you have something in mind. It's no longer a question. It's not a subjective decision. Oh, do I still want to hold on to that? Am I sure that that's still something that deserves to be in my portfolio? You've already said ahead of time, if XYZ happens, I'm going to do this, and you just do it.
That allows us to not make an emotional reaction to what we're seeing. It allows us to make some more disciplined, emotionless reaction to a set of conditions that we've already anticipated. So, I think human investors can be very effective, is having the creativity to think about what are all the things that could happen wrong for my portfolio, and what am I going to do about that and lay it out ahead of time.
The words that I rarely hear investors mention that they should mention way more often than they do is I don't know or I'm not sure. And I think I'm not sure is the more painful of the two. Because if we think we know something, we all of a sudden decide we absolutely know something. And if there's one thing that is absolutely true, is we don't absolutely know anything about investments, no, we just don't. Everything is based on probabilities and never certainties.
The reason why we have so much trouble with that is because we are hardwired as humans to want to have certainty. We want to feel that experts know things that are unknowable. And we want to feel that there is an investment process which will be at perfection and won't have any issues. But if you've traded or invested one day, you'll know that that's not the case. Things are always based on probabilities. And the best thing you can do is set yourself up for a probabilistic set of outcomes. It's never for what's definitely going to happen.
The time recently, when that really hit me was the first time I've done better and better going on financial media, on television and online TV and things like that. But the very first time I did, I went on and I gave a very honest investment approach. I said, I'm really not sure what's going to happen. But I could see this happening. And I could see this happening. And if x happens, I would bet on this. And if y happens, I'll go on ahead.
And you know it was fine as about a five, 10-minute interview, like right on the way to the elevator, he said, it would be great if you could just be more certain about exactly what you're expecting. And I'm thinking to myself, okay, and I got on the elevator. I'm thinking, well, that makes sense. But I'm not certain and no one really is, but we need that certainty.
So, when I go on television and I have three minutes to pitch an investment thesis, you have to imply a certainty because you have a limited amount of time to drive home a soundbite, drive home a thesis. But if I'm really trying to manage a portfolio or coaching my clients to look at asset allocation, it's never based on certainties, you really don't know the answer.
And I would say one of my mentors used to say, Mike Epstein, who was a legendary trader, he was based in New York and actually was, if you know the CMT Association, he was the first one to connect the New York society with the Boston buy side technical analyst and link and turn the CMT association in the global organization it is now. But Mike always used to say them that know, know they know, and them that don't know, don't know they don't know. And the problem is, no one really knows everything about an investment approach.
So, in general, we have to accept the fallibility, the imperfection of financial analysis, and that's technical, fundamental quantitative, any of those things have an imperfection built into them, because we never going to know what the future holds. And the way you address that or the way you get past your struggles of uncertainty is to have a good game plan. Know that it's going to be imperfect and have a regular period where you review what you're doing.
So, what I coach clients is once a month, once a quarter, once a year, you have a regular period where you go back and review your wins, your losses, your best picks, your worst pics, your outperformers, your underperformers and try to pick apart how your tool kit can improve and try to minimize the uncertainty, maximize having a higher probability outcome to what you're doing.
So, often, in my discussions with clients, we're talking about keeping it simple. And I found as an industry, and as investors, we love to make things more complex. And as an industry as a whole, we love complexity. We love to make things more complicated. Because it seems more rigorous, it seems more detailed, it seems more effective, when in reality, the more simple things are, the more consistent and the more robust they tend to be in terms of the results of them.
In my former life, I used to work with a lot of the trading floors here in New York. So, I was at a trading floor downtown, and I was working with the currency traders on the FX desk. And so, I spent time walking around and my job at the time was, as a technical analyst, to walk around and make sure that- just to see how people were using charts, see how people were visually analyzing their space, and then to help provide some guidance on how they can improve it. And I noticed one trader in particular had the cockpit of a ton of screens. And as I looked, there had to be 50 or 60 different technical indicators. He literally had every technical indicator I had used was on the screen at once, all across the screen.
So, I'm watching him for a couple minutes. And I'm thinking, like, this is unbelievable, I can't believe how he's able to use all this. So, I shook his hand, introduced myself and I said I'm a technical analyst, you're using an amazing amount of information. Can you tell me like what's your secret? How do you use it? He said, I have no idea what any of this stuff means. I'm like, okay, I said, but the head of the desk walk by and go, wow, look at that guy. He's all over the markets.
So, if you want to impress people, if you want to make things look more complicated, you can do that. But I've never put a chart or an insight in front of a client that has more than a couple inputs at any time. It's really important to keep it simple. And the more information you put on a chart, the more detail you try to include, the more it ends up clouding the issue. It doesn't clarify the issue, it ends up clouding it.
So, a lot of times I'm coaching my clients, we work on simplifying, we work on simple inputs, consistent inputs that will allow them to draw consistent conclusions. Otherwise, you open yourself up to all those behavioral biases, confirmation bias and others that end up just getting in the way of good returns.
How do I keep it simple, I like to you use a combination of trend following and mean reversion. And what I mean is, I like having part of my toolkit that's helping me understand where the trends are. Defining trends, recognizing trends, and understanding when trends might be exhausted, when they might be a little long in the tooth. The other side of that is play the mean reversion which is understanding when things get overextended. So, when has the market gone too far too quickly, and might be poised for a rebound.
And in general, looking at equity markets primarily over longer term timeframes, you're better off being a trend follower. So, it's better over six to 12 months to bet on what's has been working and to underweight what has not been working. In general, that's what the data proves out. But in the short term, if you're looking at a couple days to a couple of weeks, you really should be betting on mean reversion. Meaning, you should identify those little peaks and valleys within those longer trends.
So, the sweet spot I would think in terms of keeping it simple is having an indicator, an approach that's going to help you measure the long term trend, and then having a separate part of your toolkit simplistically telling you when you might want to revisit something because it might be overextended. And if you use the two of those in combination, overall, you're going to do pretty well. You won't hit everything. There's nothing perfect but you will be on top of most things and you won't be caught off guard which is what's most important.
So, it's funny, Tom Dorsey who's a mentor of mine, a friend. He started Dorsey Wright years ago, he's a pioneer in point and figure analysis, has done some exceptional work with that. He shared a story about when he was a test pilot in the military and he was flying an airplane, southern part of the United States, and basically got disoriented. And that happens a lot as a pilot, you're trained to figure out where you are geographically, and you have to understand where your position relative to everything else. But it's very easy to lose place of where you're at, because you're focusing on a bunch of other things.
So, he was basically and it's so funny the way he described it, he was looking for the ocean. So, he'd be able to reorient himself, okay, I know, that's west. And now, I need to do this. So, what happens is, he's looking around, and he'd made a bunch of turns and everything, and all of a sudden, he sees this body of water, and he's like, well, that's got to be the ocean, that's strange, I wouldn't have thought it's over there. But that's fine. I should be able to turn left, the airport should be right there.
And he was getting low on fuel, which 90-some percent of airplane accidents are running out of fuel before you get back to the airport. That's not the situation you want to be in. So, he starts looking around, and of course, makes the turn and there's no airport. And it's a bunch of fields. And he's thinking, okay, this is bad. So, in the end, he just has to put the plane down in the middle of the field, he had somehow flown over the border, and ended up landing in the middle of nowhere in Mexico.
So, the lesson to that story is number one, it's to trust your instruments, and you have tools in front of you. And at some point between when he was in a good place with his flight plan and when he was in the wrong place, he forgot to pay attention to his instruments, or he saw something and then didn't believe what he saw. And that was the biggest problem that he had. He could have seen where he was using the instruments, but he saw a body of water and then said, no, the instruments are wrong, that's where I need to be.
So, as investors, it's so funny. We have that same situation where our instruments, our model, or our inputs, our investment approach is telling us one set of decisions we should make, but we convince ourselves to do otherwise. And more often than not, that's because of behavioral biases. So, something like confirmation bias is probably one of the more common where you decide you're bullish, I see the market move up, or recently here, the S&P 500 has dropped very quickly, that happens, okay, I'm bearish. And then, instead of paying attention to your instruments, which might not be quite too bearish, you just pay attention to the news headline, or the movement or your portfolio losing a lot of value very quickly. And obviously, you've gotten away from the discipline of your process.
So, the most important thing we can do as investors is to trust our instruments, that means our instruments have to be pretty good to start with. So, you have to make sure you're very thoughtful about what inputs you're using, and how you're paying attention, how you're structuring your process. But then we have to actually follow through with what we're seeing.
So, when do we need to change our instruments? When do we need to upgrade our instrument panel? This is actually a very, very good question. And I would say as pilots, it's funny. I've learned how to fly so far using very traditional- using stick and rudder flying and using the dial. So, this is the old school way to measure altitude and speed and all of that. But there are way better avionics nowadays. You can use GPS, and you can moderate avionic traction really, really powerful. Even in a relatively inexpensive airplane, you can have really good tools to help you understand where you're at.
And what's happened is pilots over time have learned to integrate those tools into the process. So, at some point, if you've learned the traditional way of flying, looking at these dials, but there's something better out there, there's a new tool that's going to help you better understand where you're at or anticipate a change that needs to happen, you have to have a time to review that. So, it's funny, as pilots, every year or two, you have to go up for a check ride where you have to reestablish your expertise flying an airplane in between those points. That's a time to try and upgrade your process if you can.
And working with my clients, we talked about that same idea, with some regularity. Usually, for advisors, it's about once a year, maybe at the end of the year. Hopefully, you have a week where you can look back, see where you're at, look forward and just assess where your practice is. That's a really good time to look at your dashboard, look at your toolkit. And the problem I find a lot of people do is they just try to change their process in the middle of things. And you don't want to do that.
You need to have a dashboard and then follow that for a while and gather data about how it's working and when it's not working. And when you have issues with it, or when you have a new input that might be good, you put it on a list over here. And then when you go through regular review process at the end of the year or the end of quarter, that's when you've earned the right to look at that list and see if there's something that deserves to be included. But that regular review process is really, really important to make sure we have the best tools in front of us at any point.
So I'm often asked by investors, by institutional investors, what are the common behavioral biases, what are the biases that they should be most aware of? And I would say the most important starting point to incorporating behavioral finance in your process is awareness. Having awareness of your own imperfections and seeing where behavioral biases are affecting your decisions. And my first answer to most of those is endowment bias or endowment effect.
And if you're not familiar with that, it's essentially something that you own or something that is your own, you assign greater value to it, because it's something that you own. So if you give me a Real Vision coffee mug, and that's mine, that has greater value to me than it will to someone else, because it's something that's now mine. I attribute greater value to it.
Well, Dan Off, who's a portfolio manager of Fidelity Investments who I worked with at times, he always said that we don't own stocks, we rent them. Now, by definition, you do own stocks- don't get me wrong, you are a shareholder. And there's a responsibility there, of course. But the point was more, don't treat your stocks as if you own them, don't think of them as your possessions that you have trouble getting rid of, you're renting that position, you are renting the right to participate there, which means you need to be equally ready to part ways when it's the right time to do it, if you need to move somewhere else.
What happens is we all of a sudden, will hold stocks in our portfolio much longer than we should. So a position will go against us. But we'll hold