TTU142: Roy Niederhoffer, President of R.G. Niederhoffer Capital Management

[00:00:06] Roy

When you look at a market, you can tell, “Oh, Roy is probably making a lot of money on this day or Roy is probably making a lot of money in this month,” and all of my clients eventually develop that even though they have no idea exactly what’s in my strategy, maybe even more so than with trend following, which sounds incongruous, but it is nevertheless the case. Once you get that idea for short-term trading, then it becomes very easy to keep in a portfolio. I think the burden of short-term managers is to provide clients that comfort and it’s a higher bar than just, “Oh, stocks are up so we’re long stocks, so we made money because we’re long.”

[00:00:48] Intro

Imagine spending an hour with the world’s greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world so you can take your manager due diligence or investment career to the next level.

Before we begin today’s conversation, remember to keep two things in mind. All the discussion we will have about investment performance is about the past, and past performance does not guarantee or even infer anything about future performance. Also understand that there’s a significant risk of financial loss with all investment strategies, and you need to request and understand the specific risks from the investment manager about their product before you make investment decisions.

Here’s your host, veteran hedge fund manager, Niels Kaastrup-Larsen.

[00:01:38] Niels

Hey everyone, and welcome to another edition of Top Traders Unplugged, where today, Alan Dunne and I are joined by Roy Niederhoffer, Founder and President for R.G. Niederhoffer Capital Management as part of our mini-series focusing on the one investment strategy that beat everything else in 2022, namely, trend following, short-term trading, and managed futures more broadly.

First off, Roy, it’s great to have you back on the podcast after all these years. Thanks so much for joining us today. We really have been looking forward to our conversation and I hope you’re doing well today.

[00:02:14] Roy

Thanks. It’s great to be here again.

[00:02:17] Niels

Good stuff. Now, before we dive into all the topics that we have lined up that we would love to discuss with you, I would like maybe if we could set the stage for the conversation so that the audience knows a little bit about your firm’s background. So perhaps you could share just a few highlights about the type of strategy you focus on and where the business stands today as we’ve entered 2023.

[00:02:39] Roy

Sure. Well, it’s hard to believe but this is our 30th year of doing business. As you probably know, we were one of the pioneers of short-term trading back in the early ’90s. Currently, we’re managing about a billion and things are looking quite good for us I guess right now. We’ve had several great years in a row since COVID, especially volatility has been quite high, which is our favorite kind of environment.

So we’ve had four consecutive great years and things are, I think, very good for the space in general, for trend following, for our kind of short-term, primarily contrarian trading. Things are looking good. I think in general, there’s a new appreciation for macro that I haven’t seen in a long time. So perhaps we’re seeing a rare confluence of performance for the space and performance for the overall CTA community and performance for short-term trading and our own that will be quite positive for quite some time. So it’s an optimistic time.

[00:03:44] Niels

Indeed it is, and we certainly share your optimism, Roy. Now, our conversation today, Alan and I have created a list of different topics and we alternate a little bit in terms of who takes the lead on each. So what we normally do is we will let Alan kick it off. So Alan, where are we starting off today?

[00:04:05] Alan

Thanks, Niels. Hi, Roy. Good to speak to you. Yeah, it’d be great to kick off and get a sense on, I suppose, why your firm and you chose to focus very specifically on short-term trading. Is that from the perspective of a particular investment philosophy or why? Particularly from the perspective of a lot of the managers we speak to who are more focused on medium-term trend following, curious to why you ended up focused on short-term trading.

[00:04:34] Roy

There are a few reasons for that. Part of the reason is my original background was in neuroscience and I had the idea, after a few years of trading, that I could start to quantify some of the insights that I’d encountered on the human behavior and that seemed to be quite universal. So I had the idea to unify the ideas of, say, Danny Kahneman and in general, the overall hypothesis that the structure of the human brain influences our behavior in predictable ways and that one could quantify the impact of those behavior patterns and find it in historical price data, and that this would then be an explanation for the very, very interesting patterns of non-randomness that I was encountering when I was working for my brother in the first five years of my career from ’87 to ’92 after I graduated university.

My brother had spent, already, since 1979 quite a bit of time and effort analyzing short-term price patterns. So he’d already shown me that these things were there. I guess what I did was try to find some explanations for why these patterns existed and then try to develop an overarching philosophy for where one should look to find more of these patterns. That really helped me. I like to use the analogy, it’s like when Kepler came up with the idea that the planets moved in elliptical orbits, it allowed the prediction of where another planet should be and actually discovered a planet based on the theory. Or when you’re drilling for oil, if you have an idea where, 200 million years ago, there was a river, if you drill along that river, you’ll know exactly where to drill your wells so you don’t just drill in some random pattern. So it helps us find things. That was one reason.

Another reason was a business reason. When I started in October of ’92 when I left my brothers and began coding to write the software for R.G. Niederhoffer, it was basically a world of trend followers. There was a lot of the Turtles that were there, many of whom were still around and are still doing great, and there wasn’t much that was an alternative to that. I realized that what we could do would be negatively correlated to equities in probably a different way than trend following would. So there was a business reason too. We would be distinct and also it would be methodologically interesting. So those were the two reasons.

[00:07:40] Alan

Interesting. You mentioned working with your brother, Victor. We’ve talked to other short-term managers who have referenced their initial experience there too. So obviously, I guess he was something of a pioneer in the space in terms of short-term trading. You said you took your experience in neuroscience and that was a way of helping you explain some of these patterns that you were observing.

So do you think that that kind of behavior element, the opportunity that you’re capitalizing on, is that still primarily just those behavioral biases or characteristics of traders or is it a broader set of inefficiencies that you’re trying to capitalize on now?

[00:08:22] Roy

I would say both. I would say we are specifically looking for things that create strong emotional responses in market participants. If you then look for those, that will point you in a different direction than things that say require tremendous, complicated computations let’s say but might not provoke a strong emotional reaction. My view is that if you’re going to feel something very strongly, that will make your subsequent behavior more predictable. I think that’s true, in general, in human behavior and it’s true in behavior of traders. That’s always been where we begin to look for predictable patterns that we follow.

So there’s that piece of it. I would also say more generally, one thing I think we’ve done very well is approached the problem of finding strategies in a highly scientific way to really think of this as proper science and not explaining data and not letting computers just squeeze the data and torture it until it confesses spurious correlations and non-meaningful non-causal relationships that seem to work historically and then don’t generalize and fail to predict in real-time.

[00:10:01] Alan

You mentioned being in business 30 years now, which is quite an achievement. Obviously, you’ve seen a lot in those 30 years, many ups and downs. I guess short-term trading has had ups and downs in that period too. Even in the last 15, 20 years, we’ve had a couple of periods, a couple of lengthy drawdowns. Is that just part and parcel of these strategies you would think? If these biases are prevalent in markets, should they be more persistent or is it that these biases and these opportunities are more to the fore in particular market environments when volatility picks up?

[00:10:44] Roy

I think there’s definitely overall a correlation to realized volatility in all short-term trading approaches in that the markets has got to move enough so that you can overcome your trading costs. Unless you’re just selling volatility in the options market, it’s going to be very difficult to make money if the market is flatlining and not moving at all. Certainly, I don’t think anyone in the short-term space really appreciated how much the central banks were going to crush realized volatility with QE and then all the different versions of it that transpired later.

So what was a normal cycle of high to low in cyclical volatility that persisted until 2009 became just almost a falling tide of volatility. Like a California drought, it just went on and on and on and on until 2019, with a couple of maybe ‘13 to ‘15, there were some good trends in there and an occasional bursts in ’18.

I can point to a few examples that I remember well, but basically, from especially ’17 and ’19, I think it got to the point where a lot of people just gave up on short-term trading. I think one of the great fallacies is to believe that there are no behavioral biases in systematic trading. One of the questions I get is if so many people are doing quantitative trading and everyone’s automated and everyone’s systematic, how can there be behavioral biases? Well, one of the behavioral biases is that it’s hard to continue to do systematic strategies that aren’t performing as well as other things in your portfolio.

In ’17 and 19, all you wanted to be was short vol and long equities. A lot of people gave up on their negatively correlated strategies, such as trend following and such as short-term trading, and boy was that a mistake as the last three or four years has demonstrated. Unfortunately, many of the people in short-term space were not even there to survive. They didn’t survive ’19 and ’17 and weren’t there to participate, which is really a shame because there were so many good strategies that could have made money, and many investors were not able to hold on to them for that great run from ’20 onward.

[00:13:28] Niels

Yeah, I want to dig into a little bit about how investors look at these strategies. But before I do, based on what you just mentioned here, Roy, I’m curious, what made you believe in it and continue to believe in it? Because as you said, it was such an unprecedented long period of challenges for short-term trading. So what made you think that someday this was going to change?

[00:13:59] Roy

That’s an interesting question. I guess I have tremendous confidence in seeing decades of the models succeeding. I also believe that people don’t change and I suppose having a core philosophy that says, well, this is really built into the human brain and it shouldn’t change and that it is an exogenous force that’s causing it is different from saying, well, you kind of blow with the wind and whatever the market’s doing, that’s the new reality. If you have a philosophy that you believe in and you have a reason for sticking with that philosophy, it does give you the strength perhaps to continue to follow it. So maybe we were a little stronger in not feeling the need to change, so that could have been it.

But also, it could have been that we got lucky. I don’t want to attribute greats skill to what could have been just me being stubborn or not coming up with an idea to do something else whereas other people might have said let’s buy a baseball team or something else. One thing I’ve learned not to do is attribute great skill or intelligence to decisions that also where there was a lot of luck involved. A lot of people like to give advice ex post facto and say, “I was so clever back in the day.” No, a lot of it is just luck. People that are humble enough to admit that and it may have been that I was just lucky to stay with it one year long enough to participate. So it could have been that too.

[00:15:47] Niels

I appreciate that. Now, you mentioned this thing about investors’ ability to hold on to a strategy, and I’ve referenced a paper that Cliff Asness and our friends over at AQR wrote last year, which talked about different things pertaining to trend following in particular, but also what investors want. I don’t want what I’m going to say now to suggest that that’s exactly what they wrote in the paper, but it kind of leads to the question. Because I think one of the questions they raised is, are managers becoming too concerned about the Sharpe?

But the real question, I guess, is whose Sharpe should we be concerned about? Is it the Sharpe of the manager, which of course, a lot of investors will focus on that? Or is it the Sharpe ratio of the overall portfolio of the client after they include these strategies, which arguably, should be what investors should be concerned about? I’d love to hear your thoughts about this and how you take this into account or not in designing your own strategy.

[00:16:58] Roy

Sure. Well, I think perhaps to a fault maybe more than almost anyone else out there, I have been a hedge fund that actually hedges. If I were to go back to my 26-year-old self and give 26-year-old Roy some advice, I would have said, “You know where it says start a hedge fund? Don’t listen to that word. Just be correlated to the S&P like everybody else.” We ended up with a strategy that for 25 years, I just have the numbers, we ran a minus 0.4 beta to stocks, and a minus 0.6 beta to bonds, which meant that we have like, I think the number is 9% annualized alpha for 25 years.

Very few hedge funds produce 9% of annualized alpha to stocks and bonds each for 25 years. That’s a tremendous track record. The problem is I did it with this negative beta so it became very hard for investors who weren’t able to look at it on a portfolio basis, which is what I thought they would do. I kept telling people when they invest, you have to look at this on a portfolio basis. But that’s not what they do.

One of the questions I like to ask people is, “What’s the Sharpe ratio of your health insurance?” They think for a second and they come back with, “It’s like minus infinity.” So does your investment committee think about your health insurance as a standalone investment? From an investment committee perspective, it’s possibly the worst investment you’ll ever make and you should immediately redeem from your health insurance and your fire insurance and your car insurance, and yet everyone has them. We never even think twice about it.

One of the greatest experiences and most fulfilling experiences I’ve had in my career was getting to work with a great idol of mine, Danny Kahneman, who I first encountered in university as an iconic figure in academia. But we actually got to work together on a paper, a presentation that we did at a hedge fund conference a few years ago on this exact topic. Why is it so hard for investors to maintain negatively correlated and protective strategies like CTAs in a portfolio?

As part of that, we did some simulations and we came up with a really interesting hypothesis that if you have a portfolio of 20 funds in your allocation, 19 are hedge funds and one is a CTA, they all have the same Sharpe ratio. So everybody’s got the same Sharpe but they have different correlation: 19 are correlated to stocks in the way that hedge funds are and one is negatively correlated stocks and the way that CTAs are, it turns out that the CTA appears at the bottom of the monthly sorted performance table four times as often as any other manager even though everybody has the same Sharpe ratio.

It was our view that this created a situation where the CTA gets far more attention for losing money than anyone else, and of course because of loss aversion and recency bias and all the different behavioral biases that Danny identifies so beautifully in his book Thinking Fast and Slow, which I think is the best book on trading ever written by the way, they tend not to make it year after year in an allocation. They get redeemed. However, in terms of portfolio benefit, they’re the most beneficial. It turns out that even if the CTA has doubled the Sharpe ratio of all the other managers, they still are at the bottom more than anybody else. So it’s really a problem and it has to do with the skew of the strategy. Negative skew strategies are much easier to have in a portfolio than positive skew strategies like CTAs.

People love negative skew strategies because they all lose together. One of the other ideas that we came up with was based on the idea of Danny Gilbert. He used to give the most popular course at Harvard. He has an interesting idea. He calls it the emotional immune system. His idea is that tragic events or really extraordinary events that happen to you in life tend to trigger a mechanism where you never give them as much tragedy or as much happiness in your experience as they really have, so they’re muted. Whereas small events that are negative or positive do have the full value.

So he says the way to be happy is to have many, many small bursts of happiness. So if you think about negative skew strategies, they give you many, many small bursts of happiness and then they crush you all at the same time. Like the GFC, everybody lost together but we all kind of forget about the GFC. So many funds all lost together. Everybody kind of forgot about it, but it was horrible. CTAs are the opposite. They give you periods of time where they just lose a little bit and lose a little bit. We can really irk you.

It’s like you’re in a relationship and your spouse leaves their towel on the floor over and over and it just irks you. It makes you so upset and you can’t stand it anymore. It just makes you so mad, and then they save your child out of the burning building. But are you still married to them by the time they have the chance to do that? That’s kind of like CTAs.

[00:23:04] Niels

Now, I need to take you back in time, because as I mentioned, you were on the podcast for the first time back in 2014. Obviously, a very different environment, a very different time. So I had to go back in the archive a little bit to look, what did we talk about back then? I was reminded about something that actually I struggle with a lot. It was thanks to a paper you had written because obviously, I’m a trend follower, a longer-term trend follower.

You wrote this paper where (and I hope you don’t mind me quoting from our conversation), essentially, you wrote a paper that suggested or it certainly raised a big debate about short-term strategies versus trend and also the role that bonds have played in trend following returns. So the quote that I found was something like (this is you speaking, not me), “So this is a real problem, because instead of having this beautiful, smooth uptrend where you have the decline in interest rates and the roll yield all going in the same direction, if you’re short, you’ve got the direction of interest rates correct, so the contract wants to go down but the problem is the roll yield forces the contract to actually go up. This destroys trend. You can’t trend follow in fixed income on the short side. It just doesn’t work. So now you have to question well, if two-thirds or one-half or whatever the number is of profits of fixed income from CTAs came from being long fixed income and now it’s not going to work, well, what is going to work?”

So, of course, there was no comeback to your argument back then because we had not been in an environment where fixed-income markets had real downtrends. But we have now tried that for a couple of years. Did it surprise you how well trend following performed in a rising yield environment based on the research you had done where you had actually I think tried to reverse the data on bonds to get to your conclusion? Did it surprise you how well trend did once yields started to rise?

[00:25:31] Roy

Well, there have really been a couple of tests of that. The first one was, believe it or not, the fourth largest rally in rates was from March 10th of ’20 to March 19th of ’20. You probably didn’t realize that, but in fact, that was a 9% decline in the Barclays Global Ag total return index, and it was the fourth largest ever. But it came on the heels of the massive COVID rally, and it was not a particularly good time for CTAs with slightly down performance. It came after a big uptrend. So I wouldn’t say that that was a good example of CTAs making money.

But the big one was from January 26 ’22 to June ’22, where that index went down 14% and CTAs made 26%. So I would say I was completely and 100% wrong on that particular rally. The one big test of my idea was not successful. There was another decline of 7% from August 2nd of ’22 to September 20th of ’22 of 7%. That’s the sixth largest decline and CTAs also had a pretty good run. Now, so far, I would say that the evidence suggests that my idea was incorrect.

Now, I don’t know whether my theory is going to be true overall because what I didn’t postulate was an inverted yield curve environment. Because what I was suggesting was that the yield curve is positively sloping 97% of the time, or it had been until 2020. In an inverted yield curve environment however, to be short, you’re actually getting paid to be short, and that’s what we’ve had for the last few years. We’ve had the longest period of inverted yield curve in I think maybe ever. So in this situation, it’s hard to say whether it really is a test.

My view was that in a normal rising rate period where people have the idea that longer duration fixed income should be higher yield than the shorter duration fixed income, then it becomes very costly to be short and you’d have to pay the roll yield to be short. So that’s kind of the environment that I was talking about. I guess I never anticipated that we would have both the roll yield and the short position paying you, so it’s kind of the opposite of what we anticipated.

[00:28:29] Niels

No, I mean, that’s fine. I was just curious if this has sort of sparked ideas. I think your point is well taken in a sense that you had not put that into your assumption that the yield curves can invert. I guess they didn’t invert really until 2022, as far as I recall. But the other thing I was thinking is that when we think about (this is kind of going off script a little bit, but I’m sure we can do that) when you look at signals in general (so you’d know this much better than I do), because we use back adjusted prices anyways, the signal strength will be impacted anyways, to some extent, by the roll yield, meaning that if it works against you, probably the signal won’t be that strong so your position size won’t be that strong either.

So maybe in a sense, there’s a little bit of an adjustment mechanism in the way we structure our models to take into account that maybe not all environments are equal, so to speak. But you’re right. If we get both, yeah, definitely we should be able to make money.

[00:29:36] Roy

Well, that was my point that you wouldn’t actually get sales because the income markets have a natural tendency when the yield curve is positively sloping. They want to go up, so you wouldn’t see the actual adjusted prices go down enough to give you a short position. Now this time, this literally is the fastest tightening ever. So they gave you the trend signal despite that in 2020, I guess not in 2020 so much, but in 22, whatever the yield curve would have done, it still would have been a downtrend. Who knows what will happen in the future? So it may take 40 years for my theory to be tested or not. So we’ll see.

[00:30:22] Niels

Well, hopefully, we’ll have you back before.

[00:30:24] Roy

A big plus for the CTA side, for the trend following side, and a minus for R.G. Niederhoffer.

[00:30:32] Niels

Nobody’s keeping score like that anyway.

[00:30:34] Roy

But most importantly, it’s great for the industry, so everybody made money. We made money, trend followers made money, so that’s really good.

[00:30:41] Niels

That is exactly the essence of it. Alan, where do we go next?

[00:30:46] Alan

We’re talking about trend and we’re talking about all of these strategies as a group. But also, you mentioned on the role of CTAs in a broader portfolio and the great diversifying qualities, and obviously, short-term strategies occupy a unique space there. In some senses, as you say, they tend to do well in rising volatility periods, they can do well in trend periods too.

So if you’re an allocator looking at the range of options open to you, you have trend, short-term, but you also have tail risk-type strategies and option trading strategies, so I’m curious to get your perspective on where you think short-term trading fits vis-à-vis, say, volatility and tail risk trading. Why did you focus more on the futures trading as opposed to options trading in terms of executing that kind of long vol type exposure?

[00:31:52] Roy

Well, I think for a lot of options strategies, you want to get into those strategies when vol is low. Once the market starts to get volatile, it gets super expensive to enter those strategies. Imagine in say, I don’t know, May of 2008 trying to buy a tail risk strategy. Forget it. You’re talking about VIX of 40 or 50 or 60 by that point. There’s just no way you’re going to do anything in the options market with premiums that expensive [there’s no] tail risk strategy. You want to be short volatility at that point. Whereas a short-term strategy that has the ability to trade futures can do even better.

I think our Sharpe ratio, from May to November, was one of the highest that it’s ever been, versus I can’t imagine anybody made money being long options from maybe November of that year in the tail risk strategy. So I think they’re completely different and they have different uses. I think the duration of strategies from short-term to long-term are going to have dramatically different outcomes. Look, this month is a perfect example. I think we had, somebody told me, it was a 20-sigma drawdown in the CTA index or something like that this month. We’re having one of our best months ever and I think we’re not alone. Some of the other short-term managers are up tremendously. We’re having a double-digit up month, and probably it’s maybe the third worst month ever for the CTA index.

Another month like June of last year, I think, was similar where you had a tremendous reversal of the extreme. In that case, there was also fixed income coming up off the lows of the trend. Short-term strategy strategies can get rapidly in the other direction and can have a real diversifying impact. That can happen in equities on the downside, whatever it is. So I think it’s really important for people that have a diversity of duration in the CTA allocation and not think that diversification is not just having five different trend followers.

There can be some diversification there based on allocation, but the reality is, in a trend that’s extreme, you’re going to end up with all of your long-term trend followers fully invested at exactly the same time maximally as everybody was short fixed income at the end of February. If the trend ends rapidly, everybody’s going to lose at the same time. So there is that risk factor, and a lot of the short-term managers are able to be… In our case, we had actually proactively increased the amount of counter-trend knowing that there had been such an increase in trend following just on the basis of returns the prior year. Trend followers had made 40%. So we said, wait a minute, the trend reversals are going to be bigger so we should really go the opposite way. But that’s kind of how we think.

I think every one of those strategies, short-term, trend following, long vol, tail risk, has its own benefit. Where we sit, we have really focused on the specific tail risk qualities that our fund produces. In our case now, we are talking about being two-sided convex both for stocks and for bonds. We’ve been very explicit. We’ve always been really explicit about our tail risk qualities to a fault, I would say, more so than just about anyone else out there.

Again, I would probably go back to my 26-year-old self and say, don’t hedge. Don’t try to be negative 0.4 beta to stocks. I began investing in 1981 at a time when the stock market on a real basis was down 50% in the last 10 years and no one had made a dime in the stock market for the last 25 years. My grandfather, who had just died at age 90 or so, told me never invest in the stock market. I got wiped out in 1929. He never bought a stock again for the rest of his life. The short rate was 16.5%, the 30-year bond was 15.5% when I started investing. I started working for my brother in 1987 and six weeks later, the stock market went down 25% in a day. So I have PTSD from inflation and the stock market going down. It’s hard for me to get long the stock market. A lot of people never had those experiences that I did.

[00:36:53] Alan

So from an asset allocation perspective, when you’re engaging with investors, is that the message that okay, you’re going to be adding some CTAs and invest in a diversified set of CTA strategies, not just trend, not just short-term, and there is a place for each of them? Is that your positioning?

[00:37:13] Roy

Yes, I think so. What we’ve done a pretty good job of is being very explicit about the role that we play, why one should have us. We have been clear about the correlation that we provide, the negative beta, the negative correlation. I think we’re very clear about when we’ll make money, about when we’re not going to make money, how to explain that. Part of the problem that people have with short-term trading is it’s not as easy to explain as oh, bonds have gone up for five months in a row so I made money being long bonds. It’s a lot more complicated than that.

So we have the added burden of each month, and you know this very well from talking about this for many years, explaining to the investors why did we lose money, why did we make money, how, and where. I’ve been pretty good at giving people heuristics, I think, so they can understand the types of environments such that (you know this yourself), when you look in a market, you can tell, “Roy is probably making a lot of money on this day or Roy is probably making a lot of money in this month,” and all of my clients eventually develop that even though they have no idea exactly what’s in my strategy, maybe even more so than with trend following, which sounds incongruous, but it is nevertheless the case.

Once you get that idea for short-term trading, then it becomes very easy to keep in a portfolio. I think the burden of short-term managers is to provide clients that comfort, and it’s a higher bar than just stocks are up so we’re long stocks, so we made money, because we’re long.

[00:38:58] Niels

Can I interrupt a little bit, Roy, because I feel I need to dig a little bit deeper. You point out a couple of situations, and we normally never discuss performance on our on the podcast so these are just concepts or ideas. But you point out a couple of periods where you say, yeah, look at what happened and trend followers got hurt, etc., because of these reversals, and I fully appreciate that and we all know that. But there are also months like vol-mageddon, like November 2021, I think, where there was a big Omicron event.

I don’t know what the short-term index did generally speaking, but I do know some managers got hurt during those periods. Can you help me understand a little bit then, how are these months different so to speak? Because there you have also big reversals and lots of volatility expansion. How are they different from some of these other periods that you mentioned where things worked out really well for short-term, not so much for trend?

[00:40:13] Roy

There, I guess it really depends on the particular strategy and also allocation.

[00:40:18] Niels

I was just curious about it because I think Alan makes a good question about people wanting to diversify so how should they think about these different things? We kind of know the weakness of trend following. We definitely know that. As you rightly say, if bonds had been up for five months, we’re probably long, and if there’s a reversal, we’re going to lose money. Shorter-term strategies are much harder to… Well, you say your clients will know on a good day or a bad day what you’re doing but I find it really hard to predict, myself actually, if it’s going to be a good month or a bad month.

[00:40:57] Roy

I think you have to be very careful to ascertain what the drivers are of each strategy. Even within the short-term space, there are people who are doing 5 to 10-day trading and people that are doing 0 to 2-day trading, and there’s an enormous difference in those two strategies. There are short-term managers who are like 0.5 to 0.7 correlated to the CTA index, and others who are 0.2 to 0.3, like us. That has to do with duration and the asset mix.

Also, there are short-term managers who are more long equity, long fixed income for various reasons and have gone in the multi-strat-ish or short vol direction, and others who are not. So, those are some other factors. There, you’re going to see things like if there’s a massive trend reversal in fixed income, if it’s a long-term trend reversal and the manager has a longer duration, I can predict that managers going to do worse. Or if there’s a long period of nothing, if the market just flatlines, probably, R.G. Niederhoffer is going to have a tougher time than so and so other manager who’s trading with a 10-day duration, because they’re not going to trade very much, whereas Roy might get chopped up a lot more.

They might be down 3% in euro, whereas the other manager might never trade because they’re just holding one trade the entire time. Whereas I might trade five times in that same three-week period where the euro traded the same price for three weeks. So understanding that gives one an insight, I think, into the strategies. There’s just a lot more diversity in the short-term space. You just have to ask more detailed questions.

[00:42:49] Niels

Exactly. I appreciate that. That’s really helpful. You mentioned this trade frequency. One of the things we’ve certainly picked up from a couple of the other short-term managers that have been in this series, but also actually some of the bigger, longer-term managers is that they spend a lot of their research time. In fact, some of the biggest ones have said, it’s the biggest chunk of what they do in research is execution. You’ve got to a good size of AUM. You are short-term. From what I understand, you are probably one of the more active ones. How do you think about and what have you learned over the years in terms of managing this part of the strategy?

[00:43:31] Roy

I’m going to disagree with that. I think it’s a talking point. I think you can talk all day about execution, I’m not saying it’s not important, but I think it’s a smaller piece. I think it’s a percent, it’s 2%, it’s not 20%. I think if your strategy works, you should be able to execute it in almost a naive way. I’m not saying a stupid way. I’m saying if you’re smart, you don’t have to be too clever, you don’t have to be ultra clever. The difference between ultra clever and smart is not going to make or break your strategy. So that’s my first piece of it.

I’ve always said that if your strategy doesn’t show up in relatively… You don’t need tick-by-tick data to do trades that are lasting weeks at a time. If you need that, you’re probably overfitting. If you’re trading for weeks, it should show up in daily data. If you’re trading for days, it should show up in hourly data. Maybe you can get a little bit more information from finer data than that. So that’s my general view.

I’m not saying execution isn’t important, but it’s not going to make or break your trading strategy. Diversity is important. Trading size is important. Alpha is important. Having a strategy that has a 1.5 Sharpe versus a negative 0.2 Sharpe, that’s important. The difference in execution is going to be 1.3 versus 1.43 or something like that.

[00:45:33] Niels

No, I think that’s a very valid point, Roy. I tend to agree with that.

[00:45:36] Roy

Again, I’m talking about short-term trading here. You can do it poorly. I’m not suggesting that a naive person is going to have that, but as long as you’re thoughtful. What I consider normal operating procedure might be the benefit of 30 years of experience, and most people wouldn’t think to do it the way we do it.

[00:46:03] Niels

Sure. Absolutely.

Alan, do you want to continue with the research?

[00:46:13] Alan

What we were going to talk about which delves into a couple of these topics is, I guess, yeah, picking up on the point about the performance profile but also the multi-strat nature of short-term trading being short-term momentum, but also contrarian, and also another topic, which is machine learning. It’s obviously been a topic that’s been very much to the fore recently. In recent years, everybody has been very enthusiastic about what it can possibly bring in, and I think that’s something that Roy has incorporated into his portfolio.

So my question is, if you go back to maybe to talk about maybe the February 2018 type environment, vol-mageddon, which was a tough month for short-term. So that’s a month where, as an outside investor, you might hope short-term would do well, but due to a various set of factors, short-term struggled. So how do you balance?

Obviously, it’s ultimately about developing the best program with absolute return, but be cognizant that with this multi-strat approach and incorporating things like machine learning, there will be months where you won’t deliver maybe that that kind of crisis alpha for want of a better expression.

[00:47:39] Roy

Yeah. I think that was almost like an event-risk situation. I think the nature of our world is there are always going to be exogenous periods. It was certainly one you wish you can go back and do it all over again. It turned out like it was the beginning of a burst of volatility and it turned out it was an up quarter for us, but I think our worst month ever in the middle there. But it was, I think, maybe a harbinger of what happened in ’20 because it was really the first time that they were starting to think about reducing the balance sheet.

[00:48:29] Alan

When you get a month like that, so more from a research perspective, is that just obviously a negative point on the distribution or do you go away and say maybe we have something calibrated wrongly within the models? What’s the process for thinking about that? Because I know this is something that you have done with your main program, that you have reassessed and reevaluated it. Does that tend to be driven by performance or a macro view or you’re reading on how markets are trading?

[00:49:04] Roy

That month didn’t change anything. We certainly wouldn’t want to make a change after that. But my macro view changed with the inflation and the clear understanding that we’re going to have met unprecedented and completely uncontrolled debt to GDP as far as the eye can see. As a result, a central bank that is unable to bring interest rates to the level that they need to control inflation, and therefore the risk of massive upside in financial assets.

Now, for a guy that’s been talking about downside protection for his whole career as I have, maybe more than anybody out there, to start talking about, “Oh, what about..?” I’m not talking about Venezuela or Zimbabwe here, in terms of that kind of appreciation. But what if we see 20% or 30% inflation because the US decides that they’re going to print money? We have a $51 trillion debt in eight years, the next administration. This isn’t 40 years or 80 years from now, it’s eight years from now. The next president, if it’s a two-term president, $51 trillion, and that’s just on the balance sheet, not the unfunded mandates of Medicare, Medicaid, Social Security, and veterans’ benefits. It’s just what’s on the balance sheet.

If we’re funding that at 5% or 6%, that’s $3 trillion. What the heck are we going to do to fund that? It means that we’re going to have deficits as far as the eye can see and it starts to become exponential. My view is a risk factor that’s never been there before is debasing the dollar, printing it, to fund it. That means we could have tremendous nominal appreciation in financial assets, not because stocks are so valuable, but because the dollar is inflated. So, in that case, as a guy who has been talking about downside protection more than just about anybody, I have to start thinking about upside. How do I keep up with the S&P?

Of course, this is a challenge for everyone in the hedge fund industry, because if you think about the HFRI Hedge Fund Index, it’s done a very good job of beating the S&P on a risk-adjusted basis, but not on a nominal basis. Hedge funds in general, not individual hedge funds, but on an as used basis, a portfolio of hedge funds as the way people use them, have rarely outperformed the S&P, but they have a much higher Sharpe ratio, which is great when there’s no inflation. But if there is inflation, if you didn’t make 8% or 10%, last year, you lost money on a real basis.

If you’re going to make half of the S&P’s return, and the S&P is going up, and half of its return is because of inflation, well, now you’re making zero. So my view is people in our business and the alternatives business are going to have to keep up with the stock market in a way that they never did before. That means you have to be willing to take fully invested and even leveraged positions long the equity markets in a way that one never did before. You can’t just take 70 130-30 positions or whatever that is and expect to keep up in an inflationary world if what I’m saying is correct. But at the same time, I can make just as good a case for massive deflation.

So my view is there’s new risks out there and I have to be very careful not to get too negative or too positive. So that’s the change that we made to our strategy to make sure that whatever tail we get in stocks and bonds, we’re going to be just as good as we’ve always been on the left tail of stocks and bonds.

[00:53:26] Alan

That’s obviously a research project driven by a macro view. Is that a kind of a different strand of research to your normal day-to-day? How are you unearthing new sources of alpha?

[00:53:41] Roy

It’s actually very easy for us. We’ve spent so many years figuring out how to be consistently downside protective for stocks and bonds, for our clients who have that exposure, that it was almost trivial to do it.

[00:53:54] Alan

Okay. As I say, in terms of the types of strategies and insights you talked about at the outset, in terms of observing behavior, looking at new trading strategies, unearthing sources of return, what’s your main process for doing that? Is it from market observation, academic insights, a combination of all of the above or what?

[00:54:19] Roy

So I’ve hired a couple of neuroscientists recently, people with actual field experience for the first time. We have a number of people with more traditional quant finance backgrounds, and I’ve been hiring a little bit outside that path more recently. So that’s one aspect. But I think a lot of it is just market experience. We spend a lot of time watching the markets and everybody that works for me spends a lot of time in the trading room running the strategy. There’s not a separation between research and trading in my firm, and I think that’s really, really important.

My head trader, Paul, who’s been with me since day one, was a former local on the COMEX and the NYMEX, and that experience really colors his management of the strategy. The same thing with me. I’ve literally traded every day since 1987. So we think like traders, not like researchers, and I believe that it’s like how can you do war strategy if you’ve never actually been in combat? Maybe that’s the best way to put it. I think it’s very easy to think about strategy without understanding tactics and get it wrong.

[00:55:51] Alan

Maybe just one final one on this before we wrap up, just that we did talk a little bit about machine learning in that context of February 18. But in general, what has machine learning brought to the short-term trading space? What do those techniques enable you to do?

[00:56:08] Roy

It’s enormous risk and enormous ways of fitting data, and a tremendous number of pie-in-the-sky strategies that are going to fail. So I think it is an incredibly dangerous tool.

[00:56:25] Alan

Have you had success with it?

[00:56:28] Roy

We’ve had tremendous success with it, but not because we let the computers analyze the data and find features of the data. We think that is really, really dangerous. The specific reason, and I’ll just give you one example, that day with Powell when the text of the Fed announcement sent stocks soaring. A reporter asked Powell the question, “What do you think about the massive rally that’s occurring in the stock market?” Powell looked at his quote machine, the quotes and said, “Wait a minute, you shouldn’t be taking that as a bullish signal. This should be extremely bearish. We’re going to continue to tighten.” In five seconds, the stock market went down 4%.

That is not something that a machine learning feature creation module is going to be able to figure out because it literally is a deus ex machina situation. You can’t program that kind of thing. Even the weather is hard enough. That’s a nonlinear dynamic system and machine learning has a tough time with that. But we have a nonlinear dynamic system that has an adjustment factor where it literally does the opposite from an external force, and so convergence is a very tricky thing for a machine learning system when you have that factor. It’s not like object recognition or speech recognition where you have basically linear discrimination between the two, when you’re trying to decide a cup versus a bowl or something like that. You have a cup and a bowl and the border between them is hard.

Here we have plus 4% and minus 4% thanks to Powell’s whim on that particular day, and that makes it really hard. So I’ve been very, very, very bearish on machine learning for our domain, except for the next steps in the machine learning process, which are really, really useful. Step 1 is very dangerous, but steps 2 through 10 are really good, so that’s where we’ve had a lot of success.

[00:58:43] Niels

I have another question, another topic that I think is relevant for all strategies, but maybe more so for the short-term space. I think the three of us all share an optimism and certainly a strong belief and conviction that these strategies need to be part of any portfolio, right? Maybe one day, people will look at those who have fiduciary responsibility and say, well, hang on, have you got no exposure to these strategies? That’s something we should talk about.

But in any event, what it may mean is that we perhaps will see some flows into our world. Even though maybe last year, people took a little bit off the table because they had to fill a few holes elsewhere in their portfolio. But I wanted to ask you, Roy, a little bit about your view on capacity when it comes to shorter-term strategies and maybe also liquidity of markets, whether you’ve seen changes, if you have any concerns, and so on and so forth.

[00:59:47] Roy

Well, I think three issues there, which is should investors have the strategy and then capacity and liquidity. So let’s try all three of those. Should investors have the strategy? I think this year and ’20 really showed the value and just the incredible possibility of correlation between public markets, private markets, stocks and bonds. If there was ever any demonstration, and look, we haven’t even seen the private markets react, we haven’t even seen the full impact on commercial real estate in all these markets. Everything is correlated. Everything.

In addition, as another first-pass approximation, everything is short volatility to about a 0.9 approximation except for a very small handful of strategies. Managed futures and you could even say short-term trading, even more so, is long volatility. So the appropriate allocation to these strategies is far higher. We are in a world right now where, as I said before, the central banks are constrained because of their inability to tighten the way they used to and ease because of the potential inflationary impact.

Therefore, these tails that they used to be able to bring in and create these cycles of mean reversion are no longer as thin as they were. So markets can really continue, and this is a tremendous case for trend following and for short-term trading. The causes of volatility are much greater than they’ve ever been before. So I think risks are abundant now far more than they’ve ever been before. So that’s our rationale for this two-sided idea, and I think for other people.

Look, liquidity is something that people should be paying for right now. The idea of a REIT closing and suspending redemptions, the idea that you somehow are getting these private deals that people are so happy to be invested in because they’re not marking on a monthly basis or a daily basis, we have to mark and then people say CTAs are too volatile. I mean, it’s just ridiculous. To think that in March of 2020, that the average private deal was knocked down 85% or 90% in the middle of March 2020 or on the lows more recently. Of course they were, but they don’t have to mark it.

The fact that CTAs were up in these periods is just phenomenal. So do we belong in these portfolios? Yes, and diversity across the space belongs in these portfolios. So the answer is unequivocally yes, much more. But what about the capacity? What if what I say is convincing, and they say, okay, here’s a trillion dollars, go? So the answer is no, it’s a limited opportunity. Especially in the short-term space, you have to match your capacity. For us, it’s going to be $3, $4 billion. That’s about it for us at our current duration. Look, we’ll find other ways of trading.

Obviously, there’s the multi-strat path for some managers. Some have kind of gone in that direction successfully, some have not gone in that direction successfully. You can always diversify in that way. Of course, then you get more correlated to stocks. Maybe you can avoid that and maybe you can avoid getting short volatility. So there are ways of managing growth that I think some have successfully achieved, others have not. Trend following obviously has more capacity. But of course then, as a lot of people discovered, it’s very easy to just get long stocks and long bonds with predictable results if stocks and bonds go down together as they did more recently.

So that’s a pitfall. It’s a trade-off to grow in that direction. I think every manager has to address that question individually and it’s a question to ask each manager. In terms of liquidity, that’s the final question you asked me. Interestingly, across the fixed income duration, it’s very different. This short end has really changed. Two-year went from thousands up to 50 to 100 up in this banking crisis. It’s really tough to execute. We don’t trade that very much. The longer end however is just where it was. We have seen not much impact there and I think the other markets have been very good.

One thing to always take with a grain of salt is when people say oh, the amount on the bid and the amount on the offer has decreased. That’s not the key metric. The key metric is relative to the volatility that you see in the market, how much does it cost to get in your position size that you want to take relative to how much volatility you’re getting? That’s the key metric, not just how much is there on the bid or offer.

Sure that’s a metric, but for a typical manager, you want to take a position and then there’s going to be some vol of that position. Can you get into that position and then is it an appropriately sized position for the volatility is the question you want to ask, not just what’s its bid size is.

[01:05:33] Niels

Yeah, absolutely. Roy, we’re coming up on a little bit more than the hour, so I’d like to just finish off with one final question which we’ve asked everyone in the series so far. In your case, I’m going to change the wording a little bit, you can answer both. But what we’re trying to find out is what’s the worst or the one thing that you have heard about short-term trading, or trend following for that matter, that you disagree with the most?

[01:06:03] Roy

I think there are some misconceptions about short-term trading. One is it’s trend following. It’s definitely not trend following. So that’s one. I think there’s a tremendous amount of diversification to be had from trend following, especially if you’re careful picking, for some short-term managers. A second one is there’s a bleed that’s like buying options. So a lot of people mistake when I say we’re long volatility, they hear we’re long premium, not we’re long realized volatility. So that’s another one.

I think a lot of people look at track records, and they don’t appreciate the ability for manager to grow and change over time. They are looking back at track records; look at your terrible performance in 2009. They’re not asking the right question, which is, how much of what you did in 2009 are you still doing in a systematic strategy? The answer is probably very little. But yet, they’re very, very focused on that. So they put a lot of credence into actual track records of strategies that the managers threw out a decade ago or more.

So that’s, I think, another misconception that people have. It’s very important to get a sense for current strategies and the manager’s research process and how much you believe the managers bullshit managers versus their actual research process integrity versus their track record.

[01:07:56] Niels

Yeah, I think those were three very good points. I’m glad you mentioned all of them. On this note, Roy, we’re going to wrap up our fascinating conversation. Thank you ever so much for being on the podcast again, for sharing your thoughts and insights with us. We hope that we can do this again sometime in the future. To all of you listening in today, I hope that you were able to take something from today’s conversation onto your own investment journey. If you did, please share these episodes with your friends and colleagues.

From Alan and me, thanks so much for listening. We look forward to being back with you on the next episode of Top Traders Unplugged as we continue our deep dive into the CTA industry. In the meantime, go check out the show notes for this episode and all the other resources you can find on our website. Of course, not least, take care of yourself and take care of each other.

[01:08:44] Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released.

We have some amazing guests lined up for you and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.