← All Episodes
Behind the Ticker

Petra Bakosova

AI-Driven Tactical Allocation: The HTUS ETF

·26 min

Petra Bakosova is the CEO of Hull Tactical Asset Allocation, a firm founded by renowned options trader Blair Hull. Before joining Hull Tactical, Petra worked in institutional quantitative finance and derivatives. On this episode of Behind the Ticker, Petra joins Brad to discuss their tactical allocation ETF (HTUS), which uses a systematic, model-driven approach to dynamically shift equity exposure based on roughly 20 proprietary signals spanning macro, technical, sentiment, and valuation factors.

Blair Hull's Legacy and the Timing Model

Blair Hull built one of the most successful options market-making firms in Chicago and later sold it to Goldman Sachs. After that exit, he turned his attention to equity market timing, spending years building and refining models that attempt to identify when to be invested in equities and when to step aside. Hull Tactical is the vehicle for that multi-decade research effort.

Petra explains that the firm's model uses approximately 20 different signals that are independently tested and weighted based on their historical predictive power. These signals span macroeconomic indicators (employment data, manufacturing surveys, credit conditions), market technicals (momentum, breadth, volatility measures), sentiment indicators (investor positioning, put/call ratios, surveys), and valuation metrics (earnings yields, relative valuations across asset classes). The aggregate output produces a conviction score that tells the portfolio how much equity exposure to hold at any given time.

The fund can range from roughly 0% equity exposure (fully in cash or short-term Treasuries) to over 100% equity exposure using leverage. In practice, the portfolio spends most of its time somewhere in between, adjusting gradually as signals shift. Petra emphasizes that this isn't a binary risk-on/risk-off switch. It's a continuum that reflects the model's aggregate view of the probability of positive equity returns over various time horizons. The gradual adjustment helps avoid the whipsaw problem that plagues simpler tactical systems.

How HTUS Operates and What Makes It Different

HTUS primarily uses S&P 500 ETFs and related instruments to gain equity exposure, keeping implementation simple and liquid. When the model signals deteriorating conditions, the fund reduces equity exposure and increases allocations to short-term Treasuries or cash equivalents. When conditions are favorable, it adds equity. The rebalancing happens on a regular cadence rather than daily, which helps smooth out noise during volatile periods.

Petra is careful to distinguish HTUS from momentum or trend-following strategies. It's a multi-factor timing model that assesses the probability of positive market returns, not a system that simply follows price trends. Blair Hull's background in options pricing gives the firm a distinctive perspective on probability and expected outcomes. In options, you're always thinking about the distribution of potential outcomes and pricing accordingly. That same probabilistic thinking drives the tactical model.

One key differentiator is the fund's ability to use leverage when conditions are strongly favorable. Most tactical allocation funds cap equity exposure at 100%. HTUS can exceed that, which means it has the potential to compound faster during strong bull markets while maintaining the ability to de-risk aggressively during downturns. The asymmetric structure is designed to improve risk-adjusted returns over full market cycles rather than trying to perfectly time every move.

Where Tactical Allocation Fits in a Portfolio

Brad and Petra discussed the practical challenge of selling tactical allocation to advisors. The pitch is conceptually simple: reduce exposure before drawdowns and increase it before rallies. But every tactical model misses some calls, and the key question for advisors is whether the model improves outcomes over a full cycle compared to staying fully invested. Petra positions HTUS not as a standalone solution but as a complement to static equity allocations.

Her recommended sizing is 20-30% of an equity allocation in HTUS, with the rest in traditional index or active equity strategies. This way, the tactical component provides potential drawdown mitigation and enhanced risk-adjusted returns without requiring the advisor to bet their entire equity allocation on one model's signals. The fund can also complement buffer or defined outcome products, since it uses a completely different mechanism (dynamic exposure adjustment versus options structures) to manage equity risk.

Key Takeaways

  • HTUS uses roughly 20 independent signals spanning macro, technical, sentiment, and valuation factors to determine equity exposure, with the ability to range from 0% to over 100%.
  • Founded by Blair Hull, who built and sold one of Chicago's most successful options market-making firms to Goldman Sachs, then applied decades of research to equity market timing.
  • The fund uses S&P 500 ETFs for equity exposure and short-term Treasuries for the risk-off position, keeping implementation liquid and operationally simple.
  • Unlike most tactical funds that cap at 100% equity, HTUS can use leverage in strongly favorable conditions, creating an asymmetric return profile across market cycles.
  • Recommended usage is 20-30% of an equity allocation, complementing traditional index exposure with a systematic tactical overlay for drawdown risk management.

Listen to the full conversation on Spotify, Apple Podcasts, or YouTube.

Full Transcript

4,245 words

Machine transcribed from Brad Roth's conversation with Petra Bakosova, with speakers identified automatically. Timestamps link to that moment on YouTube. Lightly cleaned, otherwise unedited.

0:00
Brad Roth

Behind the Ticker is brought to you by UX Wealth Partners. If you're a TAMP user and you're sick and tired of the legacy technology they are run on and you want more customization and flexibility, as well as an AI-driven model marketplace, UX Wealth Partners is your destination. On top of that, they have institutional trading. So if you are an ETF issuer or an SMA provider looking for outsourced institutional trading, UX Wealth can also be your destination. So check out uxwp.com to find out all the ways UX Wealth Partners can help grow and make your practice more efficient.

0:55

Welcome to Behind the Ticker. We are live from the Ultimus Client Summit down in Texas, and I had the pleasure of catching up with a handful of people there. But today we have on Petra Bakasova. She is from Hall Tactical, and we are talking about their Hall Tactical Asset Allocation ETF, ticker HTUS. This is an active strategy using 40 different indicators in order to create a hedge fund-like product. Although the returns, I would say, are really appealing compared to a lot of hedge fund replication strategies. This is a pure tactical strategy I think definitely has merit for you to take a look at.

1:40

So we talk about the details, nerd out a little bit on some of the indicators on how the product works. But without further ado, please enjoy this episode with Petra Bakasova.

1:53
Petra Bakosova

Hey, Petra. Welcome to the show.

Read the full transcript (52 more sections)
1:55
Brad Roth

Hi, Brad. Thanks for having me. So before we get started, why don't you tell everybody just a little bit about your background and how you got into your current role today over at Hall? Sure. Sounds good.

2:04
Petra Bakosova

So my name is Petra Bakasova, and right now I serve as the CEO of Hall Tactical Asset Allocation, who is the advisor to the HTUS ETF. And funny enough, last week I actually celebrated my 10-year work anniversary with Hall. So I've been there for a long time. So if you're doing the math about my tenure and the length of the fund life, you will know that I've been there since before the fund has launched. And so if we take a step back about how I ended up at Hall, before Hall, I worked for two different trading prop trading firms in Chicago. My background was in applied math.

2:46

I got a financial mathematics master's degree at the University of Chicago, and then two trading firms later. I'm not saying I wasn't excited about high-frequency market making, but I was looking to do something a little bit different, even though Hall initially hired me to work for their prop trading market making operation. As soon as I started, there was this project that was sort of on the side. They had this ETF that they wanted to launch, and they were not sure if it was going to get approved. And then about two months after I started, we got approved. And then so things got real. Okay, so now we're launching this ETF, and they needed some people to work on it.

3:28

I was the new quant, so I was raising my hand, trying to get a little bit of profile.

3:35
Brad Roth

And then 10 years later, I'm still working on this fund. No, that's great. You and I have similar backgrounds. I did the high-frequency thing for a little bit. I don't think I've ever been so stressed out in my life because everything moves so fast, and you're constantly keeping an eye on everything. But before we kind of get into all of the stuff and all the things you do at Hall, I always like to ask people, any hobbies outside of work? What do you like to do when you're not working?

4:00
Petra Bakosova

I am a huge triathlon nut, so I could spend the next two and a half hours talking about triathlon.

4:07
Brad Roth

So please don't let me go there. I think I could do the running and biking parts. The swimming part is not for me.

4:13
Petra Bakosova

Okay, that's kind of where I am. I do not have a swim background, but I'd like to think I'm getting better, or at least that's what I tell myself.

4:20
Brad Roth

No, that's great. So let's talk about Hall as a whole. I know that there's a really unique and interesting background to the firm. So what all do you do for clients? How do you help clients? And kind of secondly, can you talk about the firm's founding?

4:35
Petra Bakosova

Okay, yeah. So maybe let's start with a little bit of our story. And I think the story will explain a lot about our investment philosophy, and it will sort of explain a lot about why the fund looks the way it looks and how we go about running it. So if we go, we're named Hall Tactical Asset Allocation after our founder, Blair Hall, who most of your listeners would probably know as a, market wizard, one of the most famous options traders of all time. But what they may not know is that before he was a famous options trader, he was a blackjack player. So his, he describes it as his first investment experience was in casinos, counting cards at blackjack tables.

5:18

And what does translate into what we do now, there are a couple principles that his team embraced. So one of the principles was make a lot of bets. They would play 60 hands every hour, and they would play, just hundreds and thousands of hands per weekend. And so bet often, bet proportionally to your advantage. So as they would count the cards, they would sort of have a better idea about what's remaining in the deck. So as the advantage grew, they grew the size of their bets. And then the third main principle was stay in the game. So don't blow up, stay disciplined.

6:00

And I think we sort of embraced all of the three principles with the firm and with the fund. So we do bet often, we rebalance the portfolio once a day. And then we do bet proportional to our advantage. And once we go more into what's under the hood, I can talk more about like how we estimate our advantage on a daily basis. And then third, stay in the game. So risk management is a big part of what we do and what we pride ourselves on. So even though like our strategy includes futures and includes options, I do think one of our key advantages is our experience in risk management.

6:41
Brad Roth

Well, first of all, I like Mr. Hall already. It sounds like a fun way to kind of get started. Um, I, I love blackjack, but I'm not a card counter. So they, they still let me in the casinos. Um, the other thing too, I really, um, I really get interested in tactical funds. I think the, the science and the mathematics behind trying to build models is really cool. So, the fund, which we're going to talk about at length, length today, HTUS has a really impressive track record. And I want to say, say that publicly, because when I looked up the fund, a lot of these long, short tactical funds don't have impressive track records. And so you guys have done a wonderful job.

7:22

So can you kind of walk us through the key components of the investment process and, um, what really

7:29
Petra Bakosova

Makes your approach unique? Um, absolutely. Uh, so I think, what we do and we're open about it, um, there has been a stigma with market timing and we're openly admitting to market timing, uh, or tactical asset allocation, if you will. Um, so basically, um, but before we do that, we start with the understanding that it is good to have a U S equity exposure. So I think this is like a pretty well accepted fact, if you will, in the investment industry. And we agree with that. So basically we collect about 40 or so different publicly available indicators. Um, and then we combine these indicators into different models and evaluate what we're thinking the market's going to do.

8:17

But on average, if our models don't give us a strong signal, either way, we will be a hundred percent invested in the S&P 500. So our co-exposure will be buy and hold. And then if we start getting signals, either bullish signals or bearish signals, we will deviate from this a hundred percent exposure. So for example, today we're 93% long and we can talk about, which of the indicators are giving us bearish signals to get to 93%. On other days, we might be 120% long. So the fund does have a pretty long leash. In theory, we could go up to 200% long and we could go, um, completely flat. We could go short.

8:58

But if you look at the history of our signals and we do publish our weights on our website, we rarely get outside of the 50 to 150 range.

9:09
Brad Roth

So I saw, um, when I was kind of mauling around your website, I saw, um, the sediment indicator on your website. So does, is that sediment indicator direct correlation of the S&P exposure that you want to hold or how does that kind of play into the investment process? Yes. Yeah.

9:23
Petra Bakosova

So what we publish on the website, this is the sentiment meter. That is basically our recommended exposure to the S&P 500. So that we'll update once a day. Once we update all of our signals, uh, with, most recent market data.

9:38
Brad Roth

Yeah. So are your signals. So let's go backwards a little bit. So you're taking, um, many different indicators. You're combining them into a single signal, or are you looking, are you combining maybe tranches of them and have multiple signals? So like, is it a multifaceted decision-making tree or do you basically have one signal that spits out that says, okay, this is the exposure we want to take? Uh, sure.

10:05
Petra Bakosova

So we can start, sort of start at the top. So if we're 93% long today, that number came out of an ensemble of four different models. So we do have four different models under the hood right now. And sometimes, we develop new models and weight them differently. Sometimes we merge two models into one. Uh, but basically what goes into the models, I've said 40 or so indicators. And broadly speaking, you can categorize them into four categories. So we would have macro indicators. These would be things like inflation, unemployment. We would have fundamental indicators. So for the S&P 500 universe, we aggregate earnings, we aggregate dividend yields or total yields when we add up dividends and buybacks, or indicators like book to price.

10:55

And then the third category is what we call anomalies. And these would be all the technical indicators, seasonal indicators, things that don't really have an economic explanation, but they have just worked for, four or five decades. And then the last category would be sentiment. So we do have some access to sentiment surveys. Uh, we use a whole bird survey. We use Ned Davis as one of our sentiment providers. Uh, we work with a firm called Market Psych, um, that aggregates both like social media sentiment and news sentiment. So that's kind of the fourth broad category of indicators. And then the reason why we don't have everything in one model, a lot of these indicators work for different.

11:39

So like, for example, if you look at some short-term change in sentiment or short-term change in volatility or some kind of a, complex interaction of a recent market move and volatility move, that may give you a really good insight about what's going to happen tomorrow. If you're looking at things like inflation or if you're looking at, some of these like valuations ratios, they offer more information about what might happen over the next six months. And if we try to just throw everything in one model, we may not get the type of, smooth and reasonable contributions that we want to see.

12:19
Brad Roth

Yeah, no, that makes a ton of sense. So, uh, just staying on this for a little bit longer are, so I would assume each four, each of those four models have, a weight given to them based on what you guys want to accomplish. And so do you ever change those weights or have you kind of gone back in history and figured out like the robustness of each signal and what you're trying to accomplish over the long-term? Or is that something that, you're continually optimizing or looking at tweaking?

12:46
Petra Bakosova

So this is an evolving process. And weighting these models is like every portfolio manager could tell you. That's kind of one of the bigger challenges because, even though some of these indicators may have 50 years of history, some of these indicators may have five. And some of these models may have, a 20-year back test. But even the back test, like you always take every back test with a grain of salt. So like then we have traded history. So, giving the constraints of the data and giving the constraints and the belief we have in the data, it's kind of a blended process. So we will calculate the optimal weights, doing some kind of a market with portfolio optimization.

13:25

And we'll calculate optimal weights using some kind of a utility function and grid search algorithms. And at the end of the day, we have an investment committee that gets together and looks at all these optimized weights versus what we have now and makes a decision about whether or not to make a change.

13:42
Brad Roth

Okay, great. So that was going to be one of my questions. Let's talk about that. So it sounds like almost everything you do is completely computer-driven. However, at the end of the day, there is an investment committee that says we are willing or not willing to maybe change the weighting of our models.

14:00
Petra Bakosova

Yeah. And I think that's, I would say like in the perfect world, eventually we will gather more and more data points and we will have more and more like true out-sample results. And the decisions of the investment committee will deviate less and less from what the model recommends. And I wouldn't say like we deviate a lot. This is more, I would say it's more of like operational risk level in terms of, okay, so are we suspecting that like one of the data sources might be bad? Are we suspecting that, one of the signals might be deteriorating? So we try not to, put in our opinions.

14:41

It's meant to be more of a, like a risk management layer.

14:46
Brad Roth

Yeah, no, I love it. And I think quant works best when you have a little bit of oversight instead of just like letting it run wild. So thanks for sharing all that with me. It's super interesting. So let's talk specifically about HTUS. we've talked about the investment process behind it, but like at a high level, what is the fund really trying to accomplish or deliver to investors? Yes.

15:09
Petra Bakosova

So basically our investment premise and sort of our goal is we say we aim to outperform the S&P 500 without exceeding the volatility of the S&P 500. And that's where we see ourselves in clients' portfolio. So almost everybody who has any kind of investment portfolio in the US will have exposure to the S&P 500. And basically what we're saying is, yes, it's good to have the exposure to S&P 500. And we also think you can do a little bit better. So maybe, take away a little piece of your S&P 500 exposure and give it to the HTUS. And you will have a very similar experience, hopefully a somewhat more enjoyable experience with our fund in your portfolio.

15:52
Brad Roth

So you kind of talking about the process, you answered a lot of my questions. But like given HTUS, we know like reading and you had said can take long and short positions. So when can you talk about the portfolio mechanics of when you are getting short and there is volatility? Like what is the fund investing in? So I don't want to make an assumption there. Are you options, futures? Like how are you setting up your short positions? Yes.

16:24
Petra Bakosova

So generally speaking, we will only hold a very few different assets. So we will have cash or cash instruments. So things like money market funds or T-bills. Then we will have the SPY ETF to give us S&P 500 exposure or E-mini futures that also gives us S&P 500 exposure. And then blended with our like core market timing strategies, we also have two options strategies. Which will invest in SPX options. And those will be anywhere from zero DTE to about a one month expiration.

17:00
Brad Roth

So is your short exposure, are you ever taking long term short exposure or meaning like you're taking a cycle bet? Or is the fund going to be short in like very little windows?

17:13
Petra Bakosova

So one of the components that goes into the ensemble is a six month equity risk premium model. So in theory, the six month equity risk premium could have a bearish view on what happens over the next six months. So we could be on average under invested for as long as that model is giving us a short exposure.

17:35
Brad Roth

So if you're sitting down with an advisor and you're trying to get exposure to HTUS and how are you positioning it? Like what is like how are you explaining it to an advisor? Because it seems complicated, but it really isn't at the end of the day. Meaning everything you do back here is complicated, but the strategy, the deployment of the strategy is quite, quite easy to understand. So how are you kind of explaining it to advisors if you're sitting down with one? And where are you advising that they kind of put this exposure and how much? And I know that varies, but just general.

18:12
Petra Bakosova

Yes. Yeah. Generally speaking, again, we're going to the advisors and basically the three use cases was actually seen from advisors for HTUS. One was, people are just looking for different alternatives to the, large cap core ETFs. People are asking, okay, is there something other than SPY I could be buying? And some advisors might be interested in offering this as an alternative. The second option is a lot of clients are interested in these quantitative strategies. There is a segment of clients who are interested in, it's in terms of like the research under the hood.

18:59

Like this is a hedge fund style strategy. Like we are closely tied to the S&P 500. So we're not an absolute return hedge fund. But in terms of the data we analyze and the techniques we use, this is a pretty sophisticated strategy. So a lot of clients are interested in that. And especially if we can offer them an ETF wrapper for ETF style fees, that is a very interesting proposal for a lot of advisors. And then as an active fund, we actually do end up paying out a lot of our or most of our gains in form of dividend. So that's both a warning for a lot of the advisors. A fair warning, if you have clients in not tax-sheltered accounts, they will be incurring potentially large dividends every year.

19:43

But for some people who are in tax-sheltered accounts or people who are seeking income, I've heard people be interested in the strategy for that specific reason.

19:53
Brad Roth

So I should have asked this earlier. Is this strategy only available through the ETF? Or do you have SMAs? Do you guys offer – do you sell research? Do you – did you have a hedge fund at one point or do you still have one? We never had a hedge fund.

20:08
Petra Bakosova

We sort of have a SMA program. We opened the SMA program to potential large clients. But our number one focus is the ETF. Got it.

20:16
Brad Roth

So with the ETF, it's one thing to have phenomenal returns, an awesome story. Like what is your strategy kind of going forward in terms of continuing to grow the fund, marketing, distribution? what are your strategies or how are you thinking about that?

20:33
Petra Bakosova

So there's a few different strategies. So one of the strategies is just telling the story. As we both know, market timing has had some stigma against it in the previous years. And we're sort of telling the stories like, okay, there's a difference between what people did in the 70s and what we're doing now. Like in the 70s, people thought you could just beat the market with one indicator. And we're saying you can't. We're saying like this is why we have 40s and we call them micro alphas because we think, okay, they give you signals. But individually, each of these signals is very weak. So that's why we need all these sophisticated modeling techniques to actually extract the information in some kind of a robust way.

21:17

So part of it is just explaining what we do and educating advisors and working with advisors so that they can explain the strategy to their clients. And sort of the second, big focus for us is continual research. So like we don't think we're done. We don't think the strategy five years from now will look the same way as it looks now. We want to add, more signals. Maybe, kind of going back to like one of our principles, bet more often. Like can we find signals throughout the day? Can we rebalance the portfolio throughout the day and generate additional returns? So, we sort of understand the limitations of this approach. As much as we're excited about how we have done, we do think there's room for improvement.

22:03

We want to get to a point where we're outperforming the S&P 500, as often as we can. So the continual research is the third approach. And then, hopefully, we just started some distribution marketing efforts earlier this year. So, we're excited about that and we're hoping to grow the AUM.

22:23
Brad Roth

So continuing to talk about like that development, have you guys started down the AI path yet? Are you considering the AI path yet? Where are you in that lifecycle?

22:34
Petra Bakosova

So we've always been sort of on the AI path, depending on how you define AI. I think I've had some conversations with a few folks at this conference. And it's really interesting because depending on who you ask, like you can get really extreme. Some people will say, okay, AI is just large language models and everything short of that is not AI. And some people will take it to the extreme and say like an Excel macro is AI. So, I'm sort of laughing because, it's about everybody's definition. But, we have been in sort of the machine learning realm pretty much since inception of the fund. We have always gone past, simple regression approaches.

23:16

And we have explored, regularized regressions, alarm, lassos. We have always looked into k-nearest neighbors models, tree models, neural nets. So we're definitely going down this path of these like nonlinear kind of black boxy models. And then, obviously some of the social media data that we get digested by our partners, I'm sure, that use LLMs to analyze their data.

23:42
Brad Roth

Yeah, it's funny. It kind of goes back to, the point you made earlier where, any given signal, there's no perfect signal, right? And there's people out there in the AI space right now who are starting to see, this wave of AI funds starting to be launched. And, some of them are just sentiment only, right? They're only looking at social sentiment and they're making bets. And some of them are, fractal and pattern recognition only. And that's only what they're looking at. So I think your approach is really interesting. I really, I've said it once and I'll say it again. the people who listen to the show go back and take a look at the track record because it is very impressive.

24:23

And so with that being said, I really appreciate your time with me. But before I let you go, where can people learn more about you? Where can people learn more about Hall and HTUS?

24:33
Petra Bakosova

Absolutely. So I would say come talk to us. And we have two websites. We have the fund website, www.halltacticalfunds.com or the advisory website, halltactical.com. We do have a somewhat active Twitter page. We've been blogging a lot this year. So we probably put out a blog about once a week. So if you want to understand a little more about, like, what's going on inside our heads, that's a good place. And then reach out. So, you'll find our contact info. Give us a call. Come see us. We're in Chicago. We're always happy to talk. We will never get bored of talking about the strategy. So we would love to talk to our listeners.

25:15
Brad Roth

Well, Petra, thank you so much for spending some time with me today. Wonderful. Thank you, Brad.