Episode #89: “Emotions Will Kill You in This Game”
Guest: Blair Hull. Blair founded Hull Investments, LLC in 1999 and currently serves as the firm’s Chairman. He was also the founder of Hull Trading Company and served as that firm’s Chairman and Chief Executive Officer. A global leader in the application of computer technology to listed derivatives trading, Hull Trading leveraged technological innovations and quantitative models to become one of the world’s premier market-making firms, trading on 28 exchanges in nine countries. At its peak, Hull Trading Company moved nearly a quarter of the entire daily market volume on some markets, executed over 7% of the index options traded in the United States, 3% of the equity options, and 1% of all shares traded daily on the New York Stock Exchange. Trader Monthly recognized Blair for having executed one of “The 40 Greatest Trades of All Time,” and Worth Magazine named him one of “Wall Street’s 25 Smartest Players.” In 2014, Blair was awarded the Joseph W. Sullivan Options Industry Achievement Award from the Options Industry Council in recognition of his outstanding lifetime contributions to the growth and integrity of the U.S. options market.
Date Recorded: 1/08/18
To listen to Episode #89 on iTunes, click here
To listen to Episode #89 on Stitcher, click here
To listen to Episode #89 on Pocket Casts, click here
To listen to Episode #89 on Google Play, click here
To stream Episode #89, click here
Comments or suggestions? Email us [email protected] or call us to leave a voicemail at 323 834 9159
Interested in sponsoring an episode? Email Jeff at [email protected]
Summary: In Episode 89, we welcome legendary market veteran, Blair Hull.
We start per usual, with our guest’s background. In this case, long-time Meb Faber Show listeners may think they’ve heard it before. That’s because Blair’s background shares an interesting similarity with that of Ed Thorp – the card game, Blackjack.
It turns out Blair made a considerable sum of money playing Blackjack after reading Ed’s writings on the game. Blair tells us you needed an advantage, and then you need to stay in the game. That’s why he played with a team. More hands played according to their system tilted the odds in his favor. This is a fun part of the podcast you’ll want to listen to for all the details, including Meb’s foray into card counting with a partner that botched the system after drinking too many Bloody Mary’s.
Eventually, Blair took his winnings and used them to get a seat on the Pacific Exchange, where he became a market maker and began trading options. Blair tells us he was intrigued with market timing, resulting in a paper he wrote which concluded that you can time the market.
Meb asks about the genesis of Blair’s market timing strategies.
Blair points back to Blackjack – each different card provides an idea about the future. In a similar way, various indicators provide an idea about a market’s future. So, part of the challenge is which indicators do you consider and what weights do you put on them?
Next, Meb digs deeper, asking for more specifics of Blair’s strategy, inquiring about the indicators.
Blair mentions one indicator that piqued his interest – the Federal Reserve Bank Loan Officer Survey. They found the correlations with 6-month returns was about 30%, which is a fairly high correlation for an indicator. He then took this indicator and combined it with a few others and ran a regression with no forward-looking bias to see if they could exceed the returns of the S&P. What were the results? You’ll have to listen.
The conversation bounces around a bit before Blair mentions how valuation is one of their key variables. He tells us his valuation method combines three different aspects: CAPE, cyclically adjusted dividend yield including buybacks, and book-to-price.
The guys spend a while discussing the various inputs in Blair’s model before discussing sentiment (which Meb calls “squishy). Both guys like sentiment, with Blair even having invested in two different firms that are using Twitter feeds so he can get a better handle on sentiment.
Next, Meb asks about AI, and how machines may affect investing going forward. Blair has a proprietary trading firm that operates on a high frequency basis, so he gives us his thoughts, noting that a key to maximizing wealth is to use an optimal-sized bet.
Meb changes direction, asking what Blair is excited about today.
It turns out Blair is focusing on the stigma of market timing. He believes it will be irresponsible not to be involved in market timing over the next 30 years. That’s because when we have correlations that really go to “1” when we have a disaster, getting an edge in the market is critical.
There are a couple quick questions – Blair’s favorite indicator, and Blair’s advice to young quants looking to get into quant finance today, but then we turn to Blair’s most memorable trade.
This is a great one involving the crash in ’87, when Blair was a market maker. Don’t miss it.
There’s plenty more in this great episode featuring a true market legend, including why Blair tells us “Emotions will kill you in this game.”
That and far more in Episode 89.
Links from the Episode:
- 00:50 – Blair’s background and his grandfather that would chart stocks by hand
- 2:37 – The steps that led Blair to the investing world
- 3:07 – Ed Thorp and his papers on Blackjack
- 4:25 – Ed Thorp Podcast
- 5:07 – Differences between being part of an investment team vs going out on your own
- 7:18 – Transition to option trading
- 7:47 – Stock market timing and gambling
- 7:59 – Stock Market Logic: A Sophisticated Approach to Profits on Wall Street – Fosback
- 9:45 – How Blairs think about market timing
- 10:20 – “A Legend Passes” – Faber
- 10:55 – “Return Predictability and Market-Timing” – Hull, Qiao, Bakosova
- 11:42 – The Decomposition Signal (Bottom of every daily report)
- 12:04 – Overview of the signal decomposition theory
- 15:14 – The real-world positioning of a portfolio created by Blair’s market timing
- 16:47 – How the inputs shift into the model
- 18:42 – The last time the model flashed a negative indicator and how sentiment plays into things
- 19:12 – Meb’s tweet on stock allocation
- 20:01 – American Association of Individual Investors
- 20:32 – “When It Gets Good on Main … It Often Goes Bad on Wall?” – Paulsen
- 21:39 – Baker and Wurgler Closed fund index
- 23:17 – The role of AI and machine learning in Blair’s work
- 25:02 – How to bet size when there’s not enough data
- 25:59 – Do retail investors stand a chance against the machines?
- 28:09 – Apply the strategy to foreign markets and other asset classes?
- 29:00 – Allocate Smartly, Quantopian, and tons more
- 29:53 – What else has Blair and his team excited today
- 31:36 – Meb on “Invest Like the Best” Podcast
- 32:39 – The Little Book of Common Sense Investing: The Only Way to Guarantee Your Fair Share of Stock Market Returns – Bogle
- 33:05 – Advice for young perspective quant professionals
- 37:48 – Most memorable investment in Blair’s career
- 40:31 – How Blair factors in major outlier events into his thinking
- 41:59 – Is there anything in the current market that surprises Blair?
- 43:28 – How to follow Blair: Hulltactical and click on “The Daily Report”
Transcript of Episode 89:
Welcome Message: Welcome to The Meb Faber Show, where the focus is on helping you grow and preserve your wealth. Join us as we discuss the craft of investing and uncover new and profitable ideas all to help you grow wealthier and wiser. Better investing starts here.
Disclaimer: Meb Faber’s the Co-Founder and Chief Investment Officer at Cambria Investment Management. Due to industry regulations he will not discuss any of Cambria’s funds on this podcast. All opinions expressed by podcast participants are solely their own opinions and do not reflect the opinion of Cambria Investment Management or its affiliates. For more information, visit cambriainvestments.com.
Meb: Welcome podcast listeners and Happy New Year. And today we have a fantastic show, featuring a true market vet who’s been in the business for decades. And over this time, he’s launched numerous investment companies, one of which became one of the world’s premier market-making firms. Eventually accounting for 7% of the index options traded in U.S., and about 1% of all NYC daily traded volume. He eventually sold that company to Goldman Sachs, and has been one of the old school market wizards. We’re honored to have him on the show today. Welcome Blair Hull.
Blair: Thank you Meb, it’s good to be here.
Meb: So Blair, I went back and was actually reading a lot of material and research and came across a great nugget that before we get into kinda the modern day investment stuff, I thought we’d hear a little bit about your background and eventually how you worked your way to investments. But, you had a grandfather that used to actually trade chart stocks by hand. Was he a Dow theorist guy? Was he a point and figure? What was the…did you remember that far back looking it at how Grandpa used to approach the markets?
Blair: Well, [Inaudible 00:01:57] is the fact that he did chart these every day. I thought he puts a high, low and close down, and I thought, “My gosh. That’s sort of… let me, there might be something there.” And so you know, I’m 12 or 13 years old at this time. I had no idea what he’s doing, other than… Well, I was sort of fascinated. So then as an undergraduate, University of California Santa Barbara. And then I later went to graduate school. Graduate school, business probably trying to figure out what he was doing during those times.
And I sort of concluded that, to the most part charting wasn’t of any value but that’s what sort of peaked my interested in the investment process.
Meb: So you had a little bit of a windy road before you started trading on your own. Can you go back in time and give us a little bit of a timeline on how you went from learning a little bit from grandpa to school and grad school, and saw a couple of different stops along the way, and eventually starting your own firm. What were kinda the stops and starts that got you to eventually in the investment world?
Blair: Blackjack was really what got me into the investment world. It was a book by Ed Thorp and he said that if you had lots of little cards out of the deck then you had more big cards and that would be 10s and aces. And if you get a Blackjack you got paid one and a half to one, so you had an advantage.
So I played Blackjack for five years. I played 50 days a year for five years. And it was that capital, and not only that capital that I was able to gain during that time of playing, but also the experience of dealing with risk and reward that got me into the investment business. I then went and leased a seat on the Pacific Stock Exchange and started trading options using the same kind of theory of getting an advantage and staying in the game but in this case, coming up with a value of each option. And then… So it was a very similar.
I consider the Blackjack an investment. In fact, William Sharpe, a Nobel Prize winner in economics defined an investment as a sacrifice of current consumption for expected future gain. Well, gambling is just a sacrifice of current consumption for expected future loss. So if you can actually turn the odds in your favour, you’re an investor. So I was really an investor when I went and played Blackjack.
Meb: Well, you know, it’s funny, we’ve had Ed on the podcast. And I actually did a stint when I was a glorified ski bum living in Lake Tahoe learning to count, and of course, this is decades later. But there’s so many parallels and wonderful takeaways. Like you mentioned, there’s risk management, there’s bet sizing, there’s learning you know, systematic approaches where emotions play a role.
I mean, one of the classic ones playing Blackjack is, I remember I sat down with my wife a year or two ago. We were at a CPA conference, and I was trying to teach her how to play Blackjack at the table. And basic strategy, and we lost I think, like, the first 10 hands in a row. And her reaction was, “I don’t get the point of this. If you’re just gonna lose, why would you play?” But there’s so many takeaways.
And it was interesting because you guys actually applied a bit of a… You joined a team and then kind of started your own teams for a while. Maybe talk a little bit about that before we get in the investment world, about kind of the differences of being a loan shark versus teams and kinda how that worked out.
Blair: The difficulty in Blackjack and in investing is that you need to get in the long run. You need to have enough individual investments that can prove out in the long run. It’s a very uncertain game investing or Blackjack. And so…and your chances, if you’re one of the…if you’re the world’s best Blackjack player in the world to go up, if you go up for a weekend and you play for 8 hours, for 16 hours to 8 hours, two days in a row, you get to 800 hands. Its 100 hands an hour times 8 hours. It’d be 1,600 hands in a day. Your chances of being a winner at the end of those 1,600 hands is only two out of three.
So if you were to join with another person, and then you were both go up and play, your chances of being a winner at the end of a weekend would not be two out of three, they’d be three out of four. So if you’re able to get in more hands it is helpful in terms of reducing your fluctuations of capital which is a key. You need to not only get an advantage in the game you play but you must stay in the game.
Meb: My one foray into trying a team was, I had a buddy who also learned to count cards. And we… I remember being in Lake Tahoe, I don’t know, Cal Neva, Bill Morrow, one of these terrible casinos. And we decided, “Let’s split up,” then had our bankroll instead of rules and he was a really good card counter. Now, after about an hour I checked in with one of my friends and I said, “How’s our buddy Chris doing?” He says, “Oh man. He’s down $1,000 all ready.” I’m like, “All ready? How’s that even possible?” And he says, “I don’t know but he sure is have a good time. He’s had about five Bloody Marys.”
So you learn that it’s not necessarily even the system problems but implementation of the systems that can cause challenges too. All right. So, you kind of evolved. You said, you know, look. You know Blackjack’s great. But what was the transition to option trading in the Pacific Stock Exchange? Was it just looking for different challenges and bigger markets? What was the kind of ideas that prompted you to move on?
Blair: I went into… I became a market-maker in the Pacific Stock Exchange with my winnings from Blackjack. And I was intrigued by market timing. And in fact, I presented a paper in, I believe it was ’81 at a conference. It was actually a gambling conference in Lake Tahoe. The conclusion on that paper is that it is possible to time the market. I was a big fan of Norm Fosback.
Meb: “Stock Market Logic?”
Blair: Yes. And I still maintain a friendship with Norm. And he believed, and I’m convinced that he’s right that…and it’s been sort of proven in the academic literature now that market timing is possible. What my paper did show is that, even if you were able to time the market, the returns on from options trading and Blackjack at that time there were opportunities in both of those areas, the risk-reward characteristics vastly exceeded market timing.
And so I actually…one of the greatest decisions I made was to not deal with market timing in 1980 or ’82 but to concentrate on buying inexpensive options and selling expensive options, and that’s what led to the sale of the firm at Goldman Sachs. So that was one of my best decisions. But back in…get it…move fast forward into 2008, I was convinced that there was some way to have a bigger exposure to the market at certain times and a smaller exposure and others that would enhance your overall return.
By that time I had a family office. It was managing really the proceeds of the sale and try to in an intelligent way. And so coming out in 2008, I said, “My God, there’s gotta be a way.” Well, it ends up that there was a turn in the academic literature about that time where most of the article said, “No. You can’t really time the market.” To saying that, there are a number of indicators that actually are predictive. And if you could put those into a strategy that you actually could get enhanced returns. And that’s what led to the formation of the ETF HTUS and the strategy that we have in place.
Meb: So let’s start the transition into some of these ideas now, because…we can probably talk about this for a while. So, you sold your firm, you started thinking about again these…some of these market-timing ideas. The market commentary like you mentioned, is littered with people talking about timing the market is impossible. And one of my favourite quotes, and I think it’s Jim Rogers, who I love Jim Rogers, but he’s always… I think he’s attributed the quote of, “I’ve never met a rich technician.”
And which is really funny because… I’d Googled that phrase to try to remember who it was, and it actually goes back to our blog, where we had posted when Marty Zweig passed away, that his condominium in New York City what was up, was for sale for like $150 million or something. And so I smiled because Marty was one of the most famous market technicians in history.
Anyway, so there’s a lot of debate. And obviously it gets very emotional when people start to talk about market timing. You’ve got a laundry list of famous people like Malkiel, etc. say, “You just can’t do it.” So talk to us a little bit about the genesis of some of these ideas. How you think about market timing in general? What’s the approach? We’ll certainly link to all these white papers that you all have written on the show notes. But give us your overall framework for how you think about an approach to market timing.
Blair: Well, I’m looking at it a little bit like Blackjack. If you have 10 different cards out there, every time you see a card, it gives you a different kind of idea about the future. And so the question is, what are the cards and what are the weights on the cards, that’s all. What are the indicators and then what weights you put on them?
And certainly, they’re not gonna come into play, they don’t come into play all the time. And if you can get ’em, so many you know when they come into play and then count the pertinent variables at that time, you’ve got a system that can be a winner. And we are very transparent about what we do.
And in fact, we have a new report on our website called the Signal Decomposition where we [inaudible 00:11:47] that come up with…use something like 45 different indicators but it only…we have variable selection in each of these models. So it selects only certain variables. And it comes down to about 14 variables that actually are in our models today, and that changes every day, and that’s on the website Top Tactical.
Meb: Yes, so why don’t you give us just a broad overview of that theory? So, for example, you know, you’re targeting U.S. stocks S&P 500 and you get a laundry list of indicators that people would be familiar with, everything from Baltic Dry Index to unemployment, to trend and housing. Maybe walk us through just kind of a basic overview of these two papers, or three I think even, that you all have written, and how you kind of approach it, and how a quant like yourself who used to be super involved and much more market-making higher frequency approaches a longer term sort of tactical system? You know, what’s the framework for how you guys went and kind of constructed this?
Blair: There was one indicator that actually peaked our interest, and that’s the Federal Reserve Bank Loan Officer Survey. And that’s a variable that does not appear in the academic literature. Well, it actually does. There is an article. It says that predicts [inaudible 00:13:02]. And it really says that if you’re tightening credit, that’s a bad sign. If you’re loosening credit, it’s a good sign.
And so that was a unique variable that we thought was significant. We found out that the correlations with 6 months returns was about 30% which is a fairly high correlation for an indicator in the market. And so we combined that. And then I went back and studied the academic literature to find out, find articles such as the price…whether the price to earnings ratio was important. Whether variance risk premium were things like the Baltic Dry Index, whether those actually did it. And I really leaned on the academic literature.
So what we did was we combined these, all these variables and we ran a…we ran a regression of those in a walk forward way with no forward-looking bias to see if we could actually exceed the returns of the S&P 500. And amazingly, by taking a position was proportional to our forecast, we were able to double the Sharpe ratio of S&P 500 and enhance the return.
So that’s the basis of this. And that started with the six-month model. And then we moved into…we found some indicators that were more indicative of shorter term, just the one month and we just…we published that. That’s actually out on SSRM that [inaudible 00:14:33]. The first paper is published in The Journal of Portfolio Management. The second one, we have not submitted yet for publication. But those two models are, I’d say 75% of the power in our indicators.
Meb: Well, and the cool part too is that Blair publishes pretty transparently. He’s like, “Hey, if you want to go replicate this, here’s how you do it, here’s all the series on line. Go for it.” And so we’ll certainly link to the data series that people can take a look at. And so, let’s talk about the actual kind of real world implementation, is that so you take a look at all these indicators and it gives you in recommended allocation. And I think the range if I recall was a 200% long down to 100% short, is that right? And that’s updated on daily basis?
Blair: That’s correct. Yes.
Meb: And so, talk to me a little bit about that, that traditional positioning. Is it something that most of the time it’s you know, 60% long or it varies highly day to day? What’s the kind of actual real world or people…if someone was following you, will they be making trades every day or is it like every few months? How’s that actually kinda play out in sort of real world implementation?
Blair: We actually do change the position every day. But it changes a very small amount. Well, today we’re 56% long. And that’s made up of…we target 80% of the [inaudible 00:15:56] the S&P 500, that’s our target. We try to make sure we have [Inaudible 00:16:00] to leave the S&P 500.
I’ll tell you one of the variables that is, I think is one of our key variables, and it’s called valuation. I mean, you’ve certainly used valuation and momentum in your strategies. Our valuation combines three different aspects. It’s cyclically adjusted. Well, first, it’s cyclically adjusted price-to-range ratio as designed by Robert Shiller. And then we also have the cyclically adjusted dividend yield but we also include buybacks in our dividend yield and then we also have book-to-price.
We combine those three variables to get a overall valuation using a technique called Principle Components Analysis. That’s one of the key variables that is in both the short-term model and the longer-term model.
Meb: And I think one of the things that people may or may not be familiar with or understand is talk a little bit about is…so you have X number of indicators. And I can’t remember if it’s 15, 25, but there’s a handful of indicators in this model. But you’re not always using all of them all the time. Could you explain kinda briefly how you go about putting together a model that may have shifting inputs into the actual model? Can you talk a little bit about that?
Blair: Well, we have a certain lookback period. And this is all published information. And it’s the six-month model, it actually looks back. It looks back 12 years. We found that by looking back 12 years of the last 12 years of history, we get the best results going forward. So it’s rerun, I think it’s rerun every month. Every month we do another…we look back 12 years and we run a regression and it selects a certain number of variables.
Now, if we look at the daily report that is on the whole tactical, we see that…I think there are five variables in that model, the Baltic Dry Index, the loan survey, new orders, new [Inaudible 00:18:00], the price variable and the variance risk premium. Those are the five variables that go into that model. And then in the same thing will occur with the one-month model.
Various risk premium, is there, also the loan survey but there’s also with the National Association of Purchasing Managers Index, we’ve also got the change in unemployment rates and the change in inflation. So each…there’s a selection criteria which in the first model is a correlation screening. You have to have a correlation of at least 10% in the past period. And the other was, it’s a stepwise criteria with the AIC criteria being the choice. So the model itself selects the variables.
Meb: And so the interesting thing about this is it’s dynamic. You know, and on top of that, you guys have continued to refine the model. Like you mentioned, you published at sixth-month and then the one-month, and it’s kinda continually updating. And it probably makes sense that the exposure right now. That’s probably signalling mid-bullishness, where the trend in a lot of the indicators are probably positive, but you’re starting to like see on the valuation side and some of the late inning bull market kinda indicators flashing. We actually just tweeted yesterday that the old AAI Study Sentiment Survey was showing that people’s equity allocations are creeping up to the highest level they’ve been since I think 2000.
But it probably makes sense. When was the last time that it’s actually flashed? And you may not know this, but flashed negative positioning where it would actually be short, would that be all the way back to the crisis or is it kinda dipped in any time in the past few years?
Blair: We’ve been operating this since 2012 with real money. So it’s been quite some time before we have had…since we’ve had a negative signal.
Meb: Yeah. Well, that’s a long bull market’ll do that.
Blair: But it’s interesting. We’re trying to look at things in a little, in more of a non-linear fashion than we did before. What it’s the American Association of Independent Investors, is that what it is?
Meb: Iindividual Investors. And you know, the sentiment has always been a tough category indicators for me because it’s often you know, coincident but it’s a sentiment can always get more and more extreme. So it’s kind of like a magazine cover indicator that we use for storytelling. It doesn’t necessarily play into a lot of the actual models we run but it’s interesting.
So for example, so that’s the AAI. That goes back to the 1980s. There’s another one called Investors Intelligence. It goes all the way back to the ’60s. And the Leuthold Group, which is based out of the Midwest, does a study that they just published again, where they look at the average on the investor’s intelligence back to the ’60s and the 10 highest years, the average across the entire year. And 2017 would’ve been the second highest average bullishness.
And usually as you’d expect the next year returns are pretty subpar. But you know, we often struggle with actually using sentiment as a factor. But I think they’re interesting to give you kinda a little colour. But yeah, the AAI one, part of the allocation is simply that investors allow their portfolios to drift with what the market’s doing. So, it makes sense they have a high equity allocation because they just sit there and it goes up over time. But maybe something for you to look into. Maybe some interesting series there.
Blair: AII, we have looked at that extensively, but it doesn’t mean that we…we’re not gonna go back and look at it again. I’d been…I’m intrigued by sentiments. And there’s series of papers by Baker and Ruggler [SP] are the most quoted. And they’ve got things like the Closed Fund Index. They’ve got a variety of indicators that are available. And it’s the stand…there’s a standard academic set that people use. And we’ve looked at this, we’ve looked at the put call ratio from the CBOE.
We haven’t found anything in those sentiment indicators. And I wish I could. Here’s how excited I am about… I made investments in two different firms that are using the Twitter feed. One of them was Ascentium that is around, and then the other one is MarketSite. I made private equity investments of these to be closer to the firms so I could get a handle.
And we actually had Ascentium in the Twitter feed, we did have an indicator we were using for some time, but for some reason the sentiment just changed when we had a Trump presidency, suddenly things sort of reversed on us. And there’s something happened with the political sentiment. And we haven’t been able to sort that out yet. If you could enlighten me at all with the use of sentiment in these indicators, I’d love it.
Meb: I struggle with sentiment. It’s a squishy sorta area to me is that it can always get to more extreme reading. So we don’t particularly use it, but it’s kinda nice for storytelling. So if you find anything, let me know. But I’ll send you the Leuthold article, it’s a pretty good study.
Blair: We’ve actually used some pretty sophisticated machine learning techniques too, and we have not been able to get a cross-validated signal that is meaningful yet. So that unfortunately I don’t have it. But I love your… I love the mom indicator. That is one of my favourites, favourite ones too.
Meb: Yeah. By the way, so speaking of Twitter is I had tweeted right before this chat, I said, “Hey, if you’ve got any questions for Blair fire ’em over.” And one of them was actually on the topic you just mentioned. He says, “You know, do you mention that AI or machine learning either will lead to an advantage you know, to some market players akin to high frequency in the mid late 2000s.” So while we’re on the topic of Twitter, I figured I’d kick that one over to you.
And ironically, the next question was also, how useful do you think extreme sentiment readings are. So we already answered that one. But what’s the kinda role that AI, I mean, for someone who used to have what? You guys used to 100, 200 employees, physicists, scientists, all that good stuff. What do you kinda think about? Do you think about that area at all or are you guys doing any research?
Blair: I still have a proprietary trading firm that is operating on a very high frequency basis. And so, with the amount of data that you have, we can use some statistics, we can use some techniques in there such as nearest neighbours or regression trees, random forest, some of these techniques. And I actually went back, got a certificate in statistical learning from Stanford to try to learn these techniques. But the problem with market timing is our data sets are so small.
We talk about 252 days a year, that’s 2,500 days a decade. We can have maybe 10,000 observations. And a lot of these techniques do require 100,000 plus observations to be meaningful. And so on, and to do cross validation in a way. So, unfortunately the sophisticated techniques do not apply as well to the market timing problem as they might to a higher frequency trading situation.
Meb: And there’s actually another follow-on question. And it’s funny, these are all right coming in line. It says, how does one bet size when using machine learning and or rolling regressions where you don’t have enough historical data to understand past odds? Do you just ignore it or what is the process there?
Blair: We have a looked at this in a variety of ways, trying to use the Kelly Criteria to say what edge do we really have and how can we maximize our wealth doing this? And it really comes down to, you’ve got to also pick a criteria that you’re willing to accept what kind of drawdown you’re willing to accept, which constrains you quite a bit.
We have actually chosen to go that we want 80% of the volatility of the S&P. So that’s sort of a way in which we have designed our bets at this point. But it is a key. It is really a key to maximizing wealth as the optimal size of your bet.
Meb: All right. So two follow-up questions to what we talked about. So one is, does the average retail investor stand a chance? So did you think, okay, well, if you’re gonna do market timing or you’re gonna start to replicate some of these ideas, should you just allocate to funds like you know, HTUS or should you try to implement on your own? Does the individual retail investor stand a chance against kind of the machines going forward? What’s your general kinda takeaway and thoughts and advice there?
Blair: Well, to do a good job at market timing and I think we’re doing a relatively good job. We have really the equivalent of three full-time people not counting me. So, that’s what it’s gonna take you. And then you have to have the infrastructure which is the computing power and the data sources. So you have to be able to lean on other parts of…we lean on certain parts of our operation that give us some of the expertise and some of the access to data.
One of the sources just recently that we’re looking into that…but there’s been a new paper that said that option open interest seems to be indicative of aggregate returns. And so we’re…in order to get the option open interest every day at a certain time, and you have to say that in this case, we’re using Chicago time, but at 2:45 our machines start to run, and they pick up all the data from all these sources so that…and at 2:50, we’ve gotta signal.
And then we, between 2:55 and…2:50 and 2:55, we make sure, we check to make sure that the data doesn’t makes sense and then that order is submitted for the close. So we get, we actually do get the S&P 500 close. We have the exact close on that.
The beauty of… Although I was in higher frequency strategies before, the idea of the market timing is that the market is so massive that you have a tremendous amount of capacity in any kind of a strategy that you use if you can come up with a technique.
Meb: And in a natural you know, being a quant like you are and like I am, you know, it’s actually funny is that, if you go back to the Cambria’s founding a lot of the original research we were doing was kind of this Conometric.
And it’s funny you mentioned Fosback because I remember sitting down with a lot of the guys at Ned Davis back in the day in the early 2000s and they mentioned the same thing. They took out that big, red book, put it on their desk and said, “This is one of the actual,” and you can tell Norm this, “This is one of the big inspirations for Ned Davis too.” And a lot of the models they built. And then the late Nelson Freeberg used to do a lot there. And so we were…we’ve always been very interested and I need to go update some of those models.
And the cool part is, there’s actually a couple of websites that have sprung up that start to track a lot of these sort of quant models one’s called Allocate Smartly. What’s the other one? Extragenic. We’ll post a link. There’s about eight of them, Quantopian. But a natural extension Blair, would be, “All right. So you’ve done this in the U.S., why not apply this lens to more breath, just like Blackjack and having a team, you know, could you apply this to foreign markets, developed, emerging, individual countries, other asset classes? You guys thought about that at all?
Blair: Oh, there’s no question that we’re dealing with a two-asset problem. You’re dealing with say 10 asset problem which is looking at not only at the foreign securities but also commodities, real estate, private equity too. And so, the ultimate strategy is one that includes expected returns in all of these asset classes and in locations. There’s no question about it. And it’s a logical extension for us to do that.
Meb: Well, good. I’m waiting on the white paper. We’ve digested all the other ones, so we look forward to foreign developed. So you mentioned a little bit about private investing. So your career’s kinda spanned a lot of different things. There’s been a high frequency, there’s kind of econometric stuff now, a little bit of private. What else has got you guys excited today?
Are you…and when you think about doing research because you’re clearly still very involved in a lot of these aspects. Is there anything else that is on your brain you’re thinking about? You guys launched a crypto currency trading firm yet? Well, what else are you thinking about these days?
Blair: Well, I think the key to our success is gonna be focused. And this is… I am really focused on this problem. This market stigma that has existed on market timing. And I believe there’s a…just as it has existed, that it was irresponsibly involved in market timing in the last 30 years. And I believe that it’ll be irresponsible not to be involved in market timing in the next 30 years.
And when we have correlations that go to one when we really have a disaster, that the idea of getting an edge in the market is so critical. And there’s so many…there’s so many things that actually indicators that are meaningful that have been proven, such as the turn of the month effect that comes in. The fact that stocks do go, tend to go up in the last couple of days of the month then…and in the first few days of next month. And why aren’t people a little more exposed to equities at that time?
Well, what we do in this is that we see, we try to get the optimal weights of how much you should bet on that turn of the month effect. And then there’s so many of these that come in that are significant, that if you could just get those right you’d have a big edge.
Meb: Do you have a pet favourite? I did a podcast with Patrick O’Shaughnessy. We did a factor draft for stocks. If you had to pick stocks based on traditional factors, it was a lot of fun. Do you have a pet favourite indicator? I’m not holding you to it but is there one that kind of you had to say, “I’m gonna pick one, this is my favourite?”
Blair: I could talk about valuation and how we look at valuation as priced earnings. I would say along with that, you know, it’s sort of a…this is a funny one, the Baltic Dry index had been around a long time, but that still cuts… meets our correlation screening. That one, and then the other one I would say, it only and really, in really extreme times is the variance risk premium, that’s certainly one of my favourites too.
Meb: Yeah. Valuation’s an interesting one because I feel like so many people misunderstand it, and misapply it. And I’ve kind of grown weary and frustrated with the media at this point to where I tell myself, I’m not gonna engage anymore conversations about valuation, but I keep seeing articles about it and it drives me a little nuts. But yeah, I mean, even we’re reading John Bogle’s, Jack Bogle’s updated new book recently. And where he came out and this, I think was in the fall. And he said, “You know, look, U.S. stocks, if you use simple valuation techniques that had been around for decades, 4%.”
And then he tends to be an optimist in that area. So, you know, and that was a few months ago, so who knows. A couple more questions, and we’ve gotta start winding down. We’ve had you…man, this has flown by. We got a lot of questions from people that you know, said, “Hey, ask Blair player a little bit about kind of being a young quant today.” So if someone’s coming out of school, you mentioned that you know, even you were going back to get updated on new techniques and ideas to understand the fields.
What kind of advice would you give a young grad getting into quant finance? I mean, there’s some people that said you know, as some techniques you know, as the world evolves and some techniques fade away and the competition comes too high with high frequency, what advice would you give you know, a new UCSB grad that’s coming out of school? Any general thoughts?
Blair: Well, I’d say one thing, there’s so many opportunities now with Coursera or there’s so many different courses out there. It behooves you to be in one of those courses all the time almost. Techniques are coming up so fast that there’s no question about that. And I did tend to be a little bit of a… We sort of favour the R language to either R or if you are in a very high frequency stock that’s going to be more in Python. But you’ve gotta be proficient in one of those languages. If you can’t be proficient in some coding language, you’re pretty much cooked these days.
Meb: It’s interesting because we like to call ourselves quant lights over here. So anytime I hit my head on the ceiling, on coding and projects, we ring up our buddies that in academia like Wes Gray, and say , “Wes, you gotta do this for us.” But there’s so many ideas and extensions that I think are still useful that would be a lot of fun. And I heard… I think it was… I wanna say it was Josh Brown that said, “You know, as the world gets so sophisticated and there’s so many PhDs attacking problems, and the markets get more efficient,” he said, “one of the, you know, benefits is being longer-term focused, you know, where people can still…you’re not necessarily competing on the second or day interval but focused on kind of the longer term.”
And by the way, Coursera listeners is a online education module. So Blair, maybe we’ll convince you to do a course. What’s the topic gonna be on building the worlds market timing system?
Blair: It’s could be the equity risk premium.
Meb: I think you should do it. I think that’d be a lot of fun. You’d get a lot of subscribers and probably some new investors, we…you know, we’d actually polled the listeners of our…some of the readers of the blog before we started a podcast for many years, we didn’t do a podcast because we thought it’d be more useful to do a high production video series but everyone voted for the podcast. But a lot of the learning for myself, and I know a lot of people are like this, is very visual.
So having something like that you know, put it in your brain. Maybe a to-do for 2018. All right. A couple more quick hits and we’re gonna have to let you go. As you reflect back, one of the main questions we asked our 2017 theme, which we need a new one now Jeff by the way, it’s 2018, was, what’s been the most memorable investment in your career? It could be trade, it could be good, it could be bad. But what’s the first thing that kinda pops in your head as you think about looking back on your investment career?
Blair: Now, I was involved in a trade in October of ’87 that certainly was a memorable trade. We had a major market downturn. And the crash of ’87 I was in the right place at the right time with a fairly large trade that… I was actually on the floor of the Exchange at that time. I would say that’s the… I only, I have that trade that nothing sort of overshadow that one.
Meb: So walk us through that though. So obviously the Black Monday, the crash in 1987, U.S. stock market got pummelled. Was the trade you were short going in, you were buying at the bottom? What was the methodology? Was it a gut reaction? Was it your models?
Blair: We were a market-maker on this Chicago Board Options Exchange. And we were one of the first firms to have a machine that would show our prices on for the crowd to see. Just as we’re being so transparent in this ETF, this exchange-traded fund HTUS, we were very transparent at that time trying to show everybody what our prices for our options were. That they’d come and buy or sell from us.
And that morning, volatility had gone to… historically, it was about 16 went all the way, it was in the 80s. It actually peaked at somewhere in the hundreds. And so we were… I was in that pit trying to keep order in the pit and keep our prices in line. But the Fed had actually tightened margins on… and the banks and the clearing firms had actually restricted every… even the market-makers were capital.
And so, and we had positions in…all over the place. And so we were trying to actually reduce our positions. So we were actually short, the Major Market Index at the Board of Trade, and I had to go over there that morning. And so I was the only member of the firm that had the full seat on the Board of Trade, so I went over and traded that. And I was trying to buy in small units, small lots to just try to reduce our position and the market continued to crash during that time.
But there was a rumour that came up, that the Chicago Mercantile was gonna close the S&P 500 futures. The New York SEC, Stock Exchange to all practical purposes was closed. They didn’t answer their phones, so you could not buy or sell stock really. So the only market, there was only… the only other market open was the Chicago Board of Trade which had a Dow look-alike. And so I was in that pit, and the broker came knowing that I was the only… I was the small, I was a buyer but buying small. Said, “Where will you buy 100 contracts?” And I was buying five lots most of the time. And I gave him a… at that time, it used to trade in nickels, like it would be $280 and $0.5 cents at $0.10 cents. In this case it was trading $290, but it was like $290 bid at $295.
So it’s bid at $290, but I see it says we’re on the $100. I said $285, a ridiculously low price. And he said, “You own ’em.” I had a big swallow. I almost choked saying, “Oh my God.” He later sold me another 50, and that ended up to be the absolute low of the ’87 crash. It later closed at something like $385. It went from $285 to $385, there was such a rally. It was at such a discount. So that has to mark my career as the trade for me.
Meb: That’s such a great one. And it brings back so many like, memories but also just thoughts on thinking about markets. Where there’s a great phrase, we’re talking about normal market returns are extreme. And thinking about these outliers and crazy events. I mean, looking back over the past year, I don’t know anyone that would predict that for the first time in history, particularly after the election, that the stock market would be up every single month in 2017.
And so markets are always confounding people. How does that play into your thinking about markets in general, where you have these huge outlier events both positive and negative? So 2017 being kind of an outlier on some of the lowest volatility on record. Was that something you think there’s anything has really changed or this time it’s different where something is maybe suppressing volatility or do you think it’s just a normal you know, that just a normal way markets are behaving?
Blair: Well, let me say, my only… I am completely data-driven. If I don’t have the data I can’t put it in my model. And we go completely objectively by the model. So, I have found that it is emotions will kill you in this game. You cannot…you just have to go with some kind of a systematic approach. At least for me, that has been successful over the years. So I’m not deviated from that.
Unfortunately, we didn’t have any data on who the president was gonna be and what was gonna happen. I wish I’d had some data on that. And I actually hope I don’t have that data in the future. So, all we can do is look at what we have, the data we have available. And I think the key is gonna be tightening of credit here. And when that occurs, that will signal the downturn.
Meb: Interesting, and podcast listeners by the way, Blair has the interesting life statistic of at least once leading Barack Obama in a Senate race. So, we have all sorts of interesting ideas. But as far as the volatility is speaking, is this environment something that you know, you kind of scratch your head? The biggest surprise to me in the last 10 years is certainly I think negative yielding sovereign bonds.
I think if you went back 10 years ago and said, “We’re gonna have a bunch of sovereigns or negative yielding,” that’s to me I would think most people would be like, “That’s crazy.” Is there something in this environment that surprises you at all or do you just kind of think this is markets as usual? This volatility is kinda just normal? So, is there anything at all where it makes you kinda scratch your head a little bit?
Blair: We certainly had an extended period of low volatility. I don’t know how you’d measure that necessarily, but the average volatility over the last year I guess you could measure it in that way. I do know that we’re complete… I’m completely certain that volatility will change, that it will vary, and that voluntarily will only where it…at the lowest point of possible volatility. I can’t tell you that. We just don’t know when it’s gonna occur.
Meb: Oh, that’s what makes this game fun, right? Is it’s never a dull day. You’re a former Chicago guy. So we used to call this the Jay Cutler Bull market, the former Chicago Bears QB where it’s just kind of a…no one seems to have that euphoria yet. Although that seems to be changing in the past few months, you’re starting to see a little bit of the excitement but most of it happening not necessarily in equities but in the crypto world which I…there’s a lot of banana stuff going on. Blair, it’s been a blast. Where can people… we’ll post links in the show notes, but where can people find more information on you, your fund, what’s going on, your updates, your white papers, all that good stuff?
Blair: I think you should go to hulltactical.com and look at the daily report. And then try to go down to the decomposition of the signal, which will give things like Sell in May, we haven’t talked about that. Are there any… what are your feelings about Sell in May?
Meb: That’s an interesting… There’s so many sort of indicators and ideas. And the challenge for me is always to take a step back like 10,000 feet. So for example, you know, I understand why I can kinda make an argument for valuation or for momentum. And Sell in May, I can make a story around it. And it historically has had great returns. And listeners, by the way, that’s most of the market returns tend to occur in that, what is it? October through end of April period. And even more outsized in some areas like biotech or tech, and you could make arguments about why that’s the case.
I don’t know that you know, I would ever put… It would be an indicator I would use as a potential in an ensemble. I don’t know that I would ever put a big wager on it, but it seemingly makes sense to me.
Blair: Yeah. There’s no…you’re not gonna put any…on any one variable, you’re not gonna put a majority of your assets, there’s no question about that. The only free lunch in financial markets is diversification.
Meb: Yeah. So I mean, but that’s one, you can construct a story around it and there’s others. I mean there was a humorous study that I haven’t updated in years, but it was…it showed that the vast majority of stock market returns came when Congress was out of session. And there actually used to be a fund, a mutual fund that was based on this, and I think its downfall was that it charged an obscene management fee, but they also used to staff the booths with a former Miss America pageant winner.
That fund by the way may still exist. It was called something like the Congressional Fund. We’ll have to dig that up and look into it, I don’t know. So there’s all sorts of ideas that are interesting, that may have some inputs that you could definitely construct a good narrative around. But I’m always a trend guy. I think I always default to my favourite indicator, which is just trend. Blair, look, it’s been a blast. We could do this for hours. I would love to check in again sometime when you guys start to crank out new papers. Thanks for taking the time to sit and chat with us today.
Blair: Good to talk to you.
Meb: Listeners, you can always find more information at, mebfaber.com/podcast. We’ll post the show notes, all the links to Blair’s material, white papers, websites, ETFs, books, everything else. You can always leave a review on iTunes. Subscribe to us on Castro, Overcast, any of the other good apps. Happy 2018. Good investing.