Return Dispersion: The 2025 Story | Systematic Investor | Ep.380 [REUPLOAD]

Summary

• Non-Correlated Assets: Panel highlights a multi-year surge in assets flowing into precious metals, crypto, hedge funds, and structured products, arguing the trend is still early.
• Trend Following Dispersion: Market selection, speed, and volatility adjustment drove wide CTA performance gaps, with very slow and very fast models outperforming mid-speed approaches.
• Managed Futures: Discussion emphasizes building portfolios across style, timing, and market universes to balance dispersion and improve resilience in shocks.
• Precious Metals & Crypto: Gold benefited from allocations and central bank demand, while crypto saw substantial retail adoption, both cited as diversifiers supported by liquidity conditions.
• Structured Products & ETFs: Rapid growth in buffered ETFs and structured products is reshaping market microstructure, compressing index volatility and increasing single-name dispersion; firms like BlackRock and Goldman Sachs were cited.
• Hedge Funds & Fees: Hedge fund AUM has swelled, with multi-strats and select macro managers regaining pricing power as demand for differentiated, uncorrelated returns rises.
• Investor Education: Wealth platforms are getting smarter on futures/ETFs, but allocators still struggle with randomness, time horizons, and distinguishing luck versus skill.
• Portfolio Construction: Higher rates, 60/40 correlation shifts, and capital efficiency favor greater allocations to managed futures and other diversifiers, but manager classification and robust design choices remain critical.

Transcript

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 Neil's Krup Larson. Welcome and welcome back to this week's edition of the systematic investor series with Katie Kaminsky, Jim Kasang, Rob Carver, Mark Resiminski, Rich Brennan, Alan Dunn, Nick Balters, Andrew Beer, Yov Git, and me, Neils Castrolen. As you can tell from this introduction, today and next week will be very special episodes because it's the time of the year where all 10 of us get together for one big conversation and debate. So firstly, let me start by thanking all of you for making the time for this extended recording today, which I really have been looking forward to and of course for all the time and energy you put into making every weekly episode that we produce and have published this year. It means a lot to me and based on the feedback we get from the audience, I know it means a lot to our community. We are recording on December 17th and this conversation will be split into two parts and published on December 27th and January 3rd. We have a great lineup of topics that all of you shared beforehand. And just to mention some of the themes that we will be covering during our conversation this week and next, it will be areas such as the impact of market selection on trend following, the massive bull cycle in non-correlated investments and how 3 years in it's still in its infancy. U we're going to be discussing the period 2023 to 2025 which has been characterized by large dispersion between trend managers. We're going to be discussing whether model designs are ever obvious within our world. We're also going to be talking about how long or how much data is practically sufficient time for allocators to make inference about manager performance. We're going to be talking about the process stability in a changing market landscape and also if we should move to high valve versions of CTA products and also we're going to be discussing uh a little bit about draw downs whether it's better to have a a a deep or a long draw down um and so on and so forth. So as you can hear we really have a packed agenda. So let's just dive right into it. The format for our conversation will be that each of us will select a topic and then it'll be open to comment on by the rest of the group. But generally we are going to keep it a bit fluid to see how we get on. And since it's become a bit of a tradition to let the ladies go first, why don't you Katie pick your first topic and feel free to direct it to whomever you would like to comment on this. And of course afterwards if someone else wants to voice their strong views feel free to jump in and otherwise I may prompt one or two of you. But Katie over to you. >> So um it's interesting we have been doing our annual review of the year and for 2025 a couple of things have stuck out to us. I'd say that you know you've had some sizable return dispersion this year across different managers and this is definitely prec precipitated by this huge shock that we had in April around liberation day and what we found which was interesting uh we examined a couple key factors. First, market selection. Second, uh speeds of trends. And third, we also looked at volatility sizing and how all of these factors adjusted relative performance. And we found sizable dispersion across many of these themes. I'll start first with market selection. If you select the 10 largest markets, that has performed significantly better than sort of a diversified approach this year. And so that makes sense because things like gold having a sizable allocation to that was the right call. Um but that does show sort of the dispersion of depending on your market set you could have very different results this year. Secondly for speeds we found that very slow trend strategies and very fast trend strategies weathered the storm this year. So either reacting very quickly to what was going on or sort of being complacent was actually the best way to be. So around 8 months was probably the worst place that you could be uh in terms of a typical trend window. And that shows sort of how you know different choices in your time horizons is also quite important for providing very different returns especially in a year with a big shock. And finally we looked at volatility adjusting. And this is important because when you do have a big shock like uh what we saw recently in COVID, we studied our turbulence metrics and what you saw around liberation day was three days in a row of extensive shocks, not one and then a quiet period and then a reversal which is in some sense a very challenging uh price trend signal for trend following strategies. Um and what we found was that you know volatility adjustment that was faster would be able to navigate and forget uh what happened in that giant 3-day shock uh faster and weather this particular environment a little better than sort of a more sort of stable volatility adjustment in terms of how you think about position sizing. And that makes sense if you look across different years. 2022 was the opposite, right? So actually being faster would have gotten you in and out of those positions more aggressively. So I think our key uh takeaway this year is that given the extremity of the moves this year, subtle decisions in your allocation, subtle decisions in speed created very very different results. um which you know is an important point to remember that each year is very different for the space and that we'll see this commonly with trend strategies that rely on price based information to make these decisions. So I would love to ask my fellow panelists like if they have been looking at these sources of dispersion in the space and and some of the thoughts that they have about these particular factors and and how to navigate um how to think about that dispersion this year. Yeah. And I'd love to hear from from many of you, but actually let me just add one thing to that and that is I'd love to have your thoughts also on whether you think market selection uh was more or less important than than actually speed. Um because I think that that is a tricky one. So uh anyone who has uh an opinion about this feel free to jump in. >> We found market selection in our our goal. So imagine if you just trade gold and you know or oil or just and corn um you would see that those happen to be the markets that tra t trended the best this year idiosyncratically and that's not the case over a 25 year history but it is the case for this year. Yeah, Andrew, you race. >> Edible, that's I mean that's it's that mirrors the analysis that we've seen this year that from a replication perspective, particularly the concentration in markets, if you look at the performance of our strategies versus the overall industry, you know, you can see that uh for instance, when the euro spiked earlier in the year, we looked over concentrated in the euro because we only trade two um commodity two currency contracts. On the other hand, when the yen ripped um uh so when when yen started collapsed later in the year um you know our concentration it really helped um and also to reiterate the point on slowness um is that you know there is somewhat of a delay in terms of our rebalancing definitely helped us around liberation day and I used to use this analogy was that the shorter term models uh you know were almost kept taking Trump too literally that he would say something and then they would you know the markets would respond and shorter models would immediately respond um or or the ball controls would kick in uh whereas the longer term things were really much more most investors which are let's just see how this plays out a little bit. Um so I we've definitely seen that on our side. >> I would agree but I would say that we found the shorter term models actually weren't as bad which was interesting. It was more the mid-range that was actually the challenge like the 8 month to 6 to 8 month period. So I think those perhaps were overreacting more than the shorter term. Just adding to that >> we've done something very similar to what Katie was kind of saying. Um I just pulled out some some charts just to effectively comment specifically on the points. I would make maybe two points. The local assessment of what you've all heard me speaking over the last I guess few months all those Vshapes. So if I go back let's say last August or even the SVB or even obviously Liberation Day, we found the exact same thing that Katie mentioned. If you're too quick in a Vshape, you'll get some of the first part and some of the second part right. If you're too slow, you just stomach the first part. You get all the recovery. If there's something in between, it's almost like the soldiers walking on the on the on the bridge in a synchronous manner. The bridge is basically falling, right? So there's like there's a bit of spot here whereby the speed not being too fast, not being too slow gets all those Vshapes, you know, exactly as painful as they can get. So that's like the local comment I would make. So I totally agree on this one. Maybe kind of bigger scale yearbyear. We've done a very similar exercise. You know, the way we've done it. Uh it's actually quite nice if I were able to kind of even show you. We literally just ranked anywhere from like know two month to a 12 month halfife. And then we took samples of universes a very small and concentrated. Actually, we had 10 markets. I don't know if we had some sort of a uh number in mind, Katie. Um and then we had couple of mediumsized universes, but not too random. So somehow there's a bit of thought process to how we would put them together kind of representative and then a very large one like more than 80 markets and then every year we kind of ran them either by speed or by universe. So here's the gist of it. The last three years at least based on our calculations have been the only three years in the last 25 that the longest speed or the slowest was always the best. So the last three years 23 24 25 the best. Never ever in those 25 years had we had all three together. Sorry, never ever had about three years consecutive. Um and then let's say it's nine months for example. So like relatively slow being the best. And in terms of universe, a small universe was either the first or the second best in those three years. If you now go back to 2022, exactly as Katie said, so maybe I'm not really adding too much in the discussion, the fastest was the best. It happened however that still a smaller universe was was was a better one to have. So in summary I think locally you have this situation that you know the middle ground of the speeds is the worst but I would say year on year the longer you are or the longer time you are in your signals you would have done better which is the exact contrast to what we had in 2021 2022 that's it for me >> interesting yav. >> Yeah I I think what's interesting I'm not going to comment about the results but the perception from clients was very different. So uh throughout H1 the first half of the year the question that most investors were asking given Trump given the speed given the uncertainty around Trump decision-m process should you be speeding up so that is a question that we had to field throughout the year which is should you be trading faster and faster and in fact it's quite it's quite interesting that the the models that actually performed the best are actually the ones that were slower uh which I found found it very ironic >> this This is a really important question because uh we need to figure out a way to classify different trend follower CTAs because you may find that there isn't a single manager that does everything you want and the classification scheme that I use is I I call it a a three factor classification which called STM style timing in markets. So what you and you can think of this almost in three dimensions you know what's the style which could be the volatility uh response uh markets what are the number of markets you trade and timing which would be the speed and then you when you think of this as a three-dimensional continuum you could say like well do I have managers that fit within uh different speeds different sets of markets or different styles and so when I build a portfolio as an investor I don't want to have just one single type of manager in my portfolio within this CTA space. >> Yeah. And I was just going to add to that before we go on to to Jim's uh topic and that is that um I think it was Nick today shared with us the latest Quantica paper which specifically uh talks about speed but also in their conclusion they kind of end up saying things change and there's a clear difference between what worked prior to 2022 and what's worked in the last 3 years. So maybe the way forward is to do at least as some people do have dynamic um parameter selection uh so it's not just a static you know set of parameters that you use uh and that seems to me to be a logical conclusion um so anyways very interesting topic now we're going to probably go to something very different Jim what's on your mind >> well I'm going to try not to go I'm going to start today with a step back and looking at the the broad picture not just trend but um what I would consider non-correlated assets um I think one of the most important things that we have really been talking about for the last year is the massive correlated growth of non-correlated assets uh and what I mean by that is I really uh and they're not all assets let's say strategies and assets uh I think very Few people are talking about how precious metals, crypto uh and the growth of those is also very correlated to the growth in structured products right and hedge fund strategies uh non-correlated strategies and the amount of AUM flowing into all of these um I think the reality is um that we sit in a world where long assets um broadly globally is about 400 to 500 trillion dollars and uh up until About 3 years ago, assets in uh hedge funds were about $2 trillion. Um uh about that was about 2 and a half years ago. Uh those numbers precious metals uh in terms of money uh actual assets out of the ground right was about uh about $3 trillion. U this is about again 3 years ago. Uh structured products were at about $500 billion 3 years ago. And when I say structured product, by the way, I'm including ETFs that do the same thing and mutual funds and buffers and everything that essentially does those types of things. And then crypto was at about one and a half trillion. Fast forward three years, they've all doubled, tripled, quadrupled in assets. And most of them, it's a hockey stick, right? It's a very kind of slow growth or or no growth. You know, in the case of hedge funds, basically sideways for about uh, you know, six years and then all of a sudden, boom, three years later, hedge funds are at four and a half trillion. Crypto is at 4.5 trillion. Precious metals goes from 3 to 9, right? Um, uh, you know, structure products 500 billion to two trillion. And I think this is something very few people are talking about is that we're riding a incredible wave of of money flowing into this space. uh and when I say the space all of these spaces and I think the driver is all the same which is that you have $500 trillion dollars on one side and on the other side you've gone from call it uh 6 trillion to 20 trillion but the big point is there's only 20 trillion in uncorrel correlated assets relative to call it 4 to500 trillion on the other side and you could argue that these this wave of assets rolling in is still not just early ending But very nent um that that the need for non-correlation could very you know uh is very likely um going to be higher when long assets are not performing as well if they do so for some extended period of time and also as interest rates go higher and do it for some longer period of time. We've seen a move higher in interest rates which is a major driver of it. Again, obviously the breakdown in correlation between stocks and bonds in uh in 22 is a huge driver of this. Um but also I think uh you know record valuations and and equities and and a lot of assets uh increasing debt, loss of uh US hijgemony and in argue you know or control to some extent. All of these things are risks that are undeniable and I think seeing those risks and then also seeing the lack of uh bonds being able to necessarily offset equity risk is really the main driver. It's logical but again it's hard to find capacity for non-correlated assets and so I think this is probably the biggest conversation of all um how are we how is the infrastructure and the liquidity going to build in non-correlated assets? what are non-cor you know what else can be non-correlated and seen as non-correlated I think if you ride that wave now these are early adopters coming into the space though given the risks and given what we're talking about and if big if interest rates go higher if you more secularly if we see uh risks begin to develop and like a lot of people expect a longer period of underperformance of long assets um this could be a very interesting decade for not just trend and hedge fund you know non-correlated type strategies But for all of these things and anything related to non-correlated, I think that's probably the biggest issue of all. >> I think it's a great topic and as always already see three people wanted to jump in on this. So, so Rob, you you came in first. Um, go for >> Yeah, I'll be really quick just on terminology. I think be very clear. We're talking about things that are perceived as non-correlated and which >> absolutely >> recently have empirically been non-correlated. That doesn't mean to say that they won't be non-correlated, especially in some kind of market shock. And obviously um I'd be more I'd be surp I mean some of these things clearly are designed uncorrelated and like a pure tail protection strategy that buys out of the money straddles or something like that. It's probably you could probably say well that's probably been non-correlated but other things just have happened to be non-correlated and investors think they're noncorrelated or they may well not be in the future. >> Absolutely. >> Yeah. >> I think it's it's interesting because there are different type of investors who went into each of those markets. So if you look at the the crypto market, you will see a lot of retail flow into ET into crypto and if you look at gold, there is a much more uh there's much more central bank flowing into gold and you can understand the rationale in terms of the dollar and whether that's an inflation hedge. Uh and when you look at um at at uh sort of multi strategies or the hedge fund universe, that's actually more from allocators from a portfolio construction sort of behavior. So although you're listing all three items as the same actually there are different players of oh sorry I apologize four of them there's CTAs as well um we are all there are different players who go into trying to solve the problems and they each take a different path um and I I think that's actually related from us from our perspective the most exciting thing is is actually the total portfolio allocation this methodology which says I'm I want to have uncorrelated strategies in my portfolio because I'm thinking about my portfolio as a whole and how do I get myself a better portfolio? The answer is I want to get something which is actually uncorrelated to the long only the the delta one in bonds and equities. Uh and I think that is um that's a that's a that's a good thing that we're seeing um in terms of people are kind of beginning to value uncorrelated investment uh methodologies, investment paths. Andrew, you must have some thoughts on this. >> Yeah. No, look, I first I completely agree agree with all of it and I I Jeem, I agree with you that it's it's early stages just because of the sheer dollars involved. I mean, you look at the ETF world, it's a 13 trillion market with basically no I mean, you talk about things like buffered ETFs and stuff, they're still just scratching the surface. Um, that's why Goldman just paid 2 billion for Innovator. Um, but you know, I think I think what's also happening underneath the surface, which I see very very much on the wealth management side, is is people getting smarter and more educated. And you the first time I went to a major wealth management platform and spoke to the ETF research people about future about an ETF that trades futures contracts, they didn't know what futures contracts were. And I was getting questions like, you know, what happens if the two-year Treasury goes to zero? Um and now you have you know Fidelity and Black Rockck and you know great um uh podcast you just did Neil with with um with Black Rockck you know now you have basically this sort of gradual process of education and what that does is it also drives demand because now people have a skill set right and they understand it now they've been tasked with and that's that was my theory about right after 2022 was that the the the flipping correlation of stocks and bonds the fact that everything went down together the fact that bonds had a draw down that was 4x that it has been in the previous 20 years all of you it's indelible right you can never get rid of that every asset allocation model for the next 20 years has to incorporate that the same way we still talk about NASDAQ as though it can go down 80% tomorrow because it happened to have happened in the in the 2000s um so I I agree I think there's the seismic shift going on where there's now a bucket and a lot of the challenge that people are facing is they they now have a bucket now they've had to figure out how to fill it >> yeah Alan thanks yeah Just to echo Rob's point, I think that's really important that there is that distinction between assets and strategies that are uncorrelated with equities in a bull market and those are uncorrelated with equities in a bare market and credit is the classic example. Um we saw this in you know 2022 most recently but but financial crisis as well but even um you have to keep in mind that you know global liquidity can drive a bull market in many assets and they can move up in an uncorrelated way. So things like crypto obviously even gold benefiting from um from liquidity conditions but then the question is you know which strategies are inherently um adaptive and have the ability to generate convexity in times of stress. So I think that's that's the next level of thinking that maybe a lot of investors don't have yet you know uncorrelation but over what period are you measuring that correlation and then looking into the mechanics of the assets or strategies and do they have that ability um for convexity. So, um, yeah, people are getting more educated, but maybe not not making that distinction yet. >> Jim, back to you. >> Yeah. Yeah. I think I think one thing that is so critical about this is just, you know, because it's early days and there's still a lot of infrastructure and product being rolled out. There's a lack of liquidity to support the amount of inflows coming in. And and uh, you know, the most scalable strategies tend to be the most simple um, and often uh, imperfect. Um uh and the more flows that those structure very kind of uh specific strategies whether they're hedge fund strategies or uh um you know or structured products um they themselves are having massive reflexive effects on the underlying assets themselves. And so I think that's a critical point too. We're seeing just this year we saw several really dramatic effects um uh from from both structured products and hedge funds and there and we're starting to see the interaction of these uh those effects as well. And I'll give one great example here. um this summer, you know, we've started to see a lot of when when liquidity goes down in markets, we're starting to see a lot of V compression because the amount of structural product issuance and the amount of uh V compression that's natural uh flow overwhelms the amount of liquidity and other flows. And so a period like the summer, you're starting to see dramatic V compression. This is the second year in a row and we're seeing it with, you know, exponentially higher amounts of V supply. That vault compression leads to dispersion because that vault compression primarily is at the index level, right? And so when you're compressing V at the index level, that means because idiosyncratic crystal exists, single list constituents of the indexes start moving away from each other. So we start to see massive uh dispersion during the summers when ball is being compressed. These are effects because of the massive growth of structure product. But it's being exacerbated because the growth in hedge fund assets particularly long short equity is also dramatically higher. And so what's happening? Well, liquidity is low. Vault compression is happening at the index level. Dispersion is happening. And where is the dispersion happening exactly in the the the the where the positioning is the biggest in long short equity. So there was a massive structural pain in long equity land this summer. And so these types of massive effects are uh uh massive growth is having dramatic structural effects in the outcomes uh underneath underneath the market. Um and and if this is early innings, you know, imagine next year, the year after, the year after here in this type of environment, these things are actually going to be the primary driver at some point of the underlying assets themselves. and and so I think a critical thing to be aware of um and and again if this thesis is right that this is all being driven by a move towards non-correlation and it's really a supply and demand imbalance between uh people needing non-correlation and the amount of non-correlation available um that is maybe the most important thing to understand in finance >> Nick back to you just one add one point I think what Jim said was extremely interesting because and I think Rob said it in the beginning studying historic typically the behavior statistically speaking of those vehicles however you want to call them strategies um and that's on paper what the correlation is but I think what we have to be very careful about is the externalities um that we would bring as allocators in assessing the statistical behaviors and eventually allocating into those strategies so like if for example strategy A and B historically are uncorrelated but today all of us end up investing into and all of us have the same objective which is I don't maximize the long-term geometric returns and all of us respond to a vic shock. We're going to reduce the exposure at the same time. So we will experience correlations not in a back test but in reality. So I think those externalities become much much much more relevant in the way that those portfolios are actually managed. >> It's interesting. So just from my perspective um I mean obviously these things non-correlation that's kind of how we started decades ago talking about why trend following would be interesting. Now people seem to come around to it and you mentioned the conversation with Black Rockck actually we also or Allan had a conversation with HSBC asset management and they specifically mention you know how they want to incorporate more hedge funds and I'm just wondering uh one is uh you know what's the trigger to why they think now is the right time um but also I wonder if this will actually change the fee compression at some point that we've seen because there might uh you know more demand for the best uh strategies to deliver this non-correlation. Jim, you raised your hand. Um >> yeah, I think uh I think the the demand is being driven not just by the desire for more non-correlation of the risk, but I think the critical part is interest rates, right? I mean, we have seen since 1982 a a a you know, almost secular line down to zero, and now we've seen a blip back up. And if you really get under the hood and think about why and you know Neils you and I have talked about this on on the macro pods etc. Uh the structural driver we've talked about for four or five years before this rally started to be clear the the pressures in the system are for higher interest rates and uh they may be able to counteract that for some time but those pressures are real and um and the benefit of these non-correlated strategies is not just non-correlation which we're talking about but a for a lot of them is is capital efficiency. Um you know the benefit of uh using options right is not just non-correlation or using part you know a lot of these strategies you can get exposure with dramatically uh less capital at stake and so if you can get the uh you know 5% yield or maybe eventually 10% yield and then uh also get the exposure that's going to drive a dramatic amount of flows to it as well. So I think the interest rate story is so important. Uh uh a lot of people don't realize that 6040 as a strategy really didn't exist until the 19 mid1 1980s and it's been adopted as like the way to invest because it's been easy because it's been scalable. But prior to 1982 the correlation between stocks and bonds is positive for for the 80 years prior. Um and um uh so so that breakdown in correlation and and also not just that but the capital efficiency that you gain I think is is what people are are waking up to um and I think critical to to have the outcome going forward. >> Absolutely Alan. Yeah, I think the shift in the correlation is one. Um performance across the space has improved. That's the second and I think uh higher interest rates are are a component of that. But to your point on the fees like we are seeing that already um particularly I would say in macro discretionary macro where there's not a huge number of you know managers with very long-term track records established uh names and they've been a able to raise their fees in the last uh one to two years and obviously we've seen the growth of the multistrats which have been able to charge you know enormous fees relative to what was historically the norm. So I think it's a supply and demand story for certain strategies with certain managers of certain characteristics. Um if they are in short supply, they're being sought after and and the the fees are reflecting that. So that's definitely a change from what we saw uh in the 2010s. >> Yeah. No, for sure. All right. Very good. Um Nick, what's on your mind? What topic would you like to uh bring to us? >> Let me go back to maybe the first topic, but you know, look at it from a very different perspective. Um I think as a group um for the last couple of years we've been discussing about kind of growing the pie and I think some success has been achieved uh in all fairness. I'm I'm sure Andre will will attest to that. Um so this kind of growing the pie uh which is the aspiration or maybe like a bit of um subtle objective h how can we contrast that to the performance dispersion like we're talking about the growth in a space that somehow shares some commonality but you know obviously within that commonality there is some dispersion when it comes to speed universe so on and so forth I think this year this was actually quite elevated so here's now my question know we all speak to investors in a variety of um I guess of ways. Is that dispersion a concerning factor or should we think of that as the alpha or however you want to call it um you know what what what what fees should pay for um you know does that go in line with the growing the pie objective does it go against it's just healthy competition I'm just very very keen to see what people think about that because in my reading this performance dispersion can also be a bit of a hurdle to grow further because then the whole selection exercise kicks off and kicks off quite aggressively and then once you see this spectrum of like know plus 15 to minus15 it can start appearing a bit more concerning specifically when you know investment committees have to approve those investments right so it's more of a question to the group how do you think that dispersion plays out for the industry as a whole >> Katie you came in first >> yes Nick I think this is a great question um I think it's both a challenge but also a positive thing I mean my view is you have obviously it's been very exciting with the growth of the ETF space and different types of products and reaching a completely different investor universe and I think it's interesting we wrote a paper about this last year about the managed future space and most of the assets are really in the institutional space and as we grow that you know it's it's going to give more investors access to things that are at least something that's different from what they have right but I do agree with you that the challenge of the return dispersion will be a big challenge and it's something that we're going to have to spend a lot more time educating investors about because the truth is like when you have a 20% difference between one product and another it you know you're going to all go through those cycles right so one year if your product is challenged you're going to be answering lots of questions and another year uh another product will be challenged when the other one isn't. I mean, we know that with that type of dispersion, there's a lot of explaining to do. Um, you'll be writing a lot of papers. That's for sure. Right, Nick? But, um, but I'd say that that will be a challenge because it's one thing to join in the industry, but when the benchmarks and the dispersion is that high, it's very hard for investors to understand sort of which of these should I be invested in. I still do think though with that challenge, a lot of investors, especially in the ETF space, they've never had this. So, I'm I'm happy about, you know, that we're actually making these type of strategies are giving them finally something different, whether it's CTAs, whether it's structured products, but it is it is going to be an education hurdle for sure. We often try to think about the investor base as a monolithical and I think there is extraordinary heterogeneity and wildly different preference functions. Um there are investors I mean so in a sense what we've tried to do is build our business around a a completely underserved market which is somebody who's building a model alloc model allocation is not an expert in manner futures but can recognize the statistical benefits of a 3 or 5% allocation to the strategy but does not want to spend a year getting up to up the learning curve to try to figure out how to evaluate you know whether that dispersion can play for or against them. Um so that's one particular client base. Um but there are others on the institutional side who you know where again you have a job structure where it's somebody's job to make that evaluation and they have they're not subject to line item constraints. They're often not subject to the same uh all-in fee pressures that you see um in in other parts of the universe. So I think I think what you'll see you know broadly is um is is what's really been happening which is that um uh you know that that within the universe you will have people who will view dispersion as a huge positive because it allows them to make you know allows them to pick somebody who did well last year or or or or beat up on somebody who didn't do well um uh and then constantly kind of re reconfigure their portfolios and there'll be others who say just give me a simple solution and you know something betalike and and and that and I I think you know what I've learned over the years is that people invest in what they like and it's you know you can you can try to come up with a you know so which is the who is the best person in the space you know who's the you can make those arguments in certain hedge fund categories like don't bet against Stanley Duckener over time don't bet against uh you know Andreas Halverson in the equity long short space don't bet against um uh uh you know Ken Griffin in the multistrat space but this space has has a a perception of a lot of randomness as we've even even started this conversation into who does well or not well in a given year. And I think I think that's the the narrative challenge is is separating when outperformance or underperformance is attributable to luck or what people the answer that people want is it's attributable to skill there about. So um actually I'd like to ask Jim something because obviously at Kai wealth you look at managers and you look at finding the best strategies to put together. Um so I'd love to hear your thoughts about kind of the the challenge that dispersion within a sector within a strategy might um you know present or maybe as as as Andrew and Katie said um the opportunity set uh if you understand the strategy well. >> Yeah. I mean obviously depends on the strategy but uh look uh there's always this balance between diversification right like uh going as diversified as possible uh which is a cheat code and we want to do that as much as possible if we do not know but if you are able to uh evaluate and and determine that there is some reason to prefer one over the other clearly you do not want to diversify to the ones that are less you know, good in that scenario. And and that that's a line we're always w uh, you know, walking. We um but but there's a lot that we don't know and we we don't over estimate what we know. We we we are um and more times than not we're looking to really spread out that allocation um when we really have no uh strong confidence. Um there are things uh but we're fortunate to be in a position to evaluate better than others. And I think being able to um have that edge is important. um on the margins but but uh you know again there are very few things that are free in this world and diversification is the one thing that if you can you know that that is free um and and I think that's that's where we tend to lean um when when we don't have that specific edge >> Katie >> yeah I think I I do want to make one point which I think agree with Nick a little bit and what I worry about is the difference between returns and investor outcomes a little bit and so I think return dispersion does impact that. So behavioral effects. So for example, you know, if you look at a return series where you buy the top manager and and sell the bottom one, you know, off and on and and don't just hold the strategy, you get a very different return profile. And I think that's what is concerning about the return dispersion because it will exacerbate behavioral effects where people sort of buy the winners that don't persist and you know and instead of just saying just allocating to the space and holding and so I think investor outcomes and index returns can be very very different and I think that's where you know the challenge just gets harder when you have more return dispersion and less explanability. Yeah, and I think that's a good point and I think we certainly seen uh over the last decade or two uh some real live cases of managers who became very popular only to see their AUM implode uh in the following years and clearly that's not a good outcome from the underlying investor. Now Andrew, it is your time. it's your turn to uh bring up a topic that you'd like to uh to discuss. >> I'll tell you. So it actually sort of dovetales with uh with um with with this discussion about dispersion because Katie and I were at at a u uh in Stockholm together at a at a round table um of managers in the space and and to me what's always fascinating is when I sit down with investors and they ask us about design decisions that we've made with respect to replication models, they're always kind of disappointed in our answers and that because our answers are usually like you Well, like, okay, what how many days do you look back at it? Well, you know, we kind of like this, but you know, if we switch it a little bit, it wouldn't matter that much. You know, which factors did you pick? You know, what are the very very best factors we could use? Well, we kind of picked these factors and but but what made us comfortable in all of this was the robustness of it. You could switch out factors, you could change out window links and and it's the reality is that again I'm not a quant by background but but um but in working with with a team and having now been in the space for quite a long time um you know the fascinating thing to me is that there's usually there are very very very few circumstances where you can say it's obvious it's obvious right and even this whole conversation to me is it's not I mean dispersion is because it's not obvious and in sitting at this at this round table somebody would throw out something you or should you have more short-term models in it? And you get two incredibly smart people on the on on different sides of it basically um you know basically taking different positions uh you know AI should you use AI for this should you use you know machine learning should like so so I think what's but I think what happens and I think part of the frustration that the end investors get is that by the time people have made these design decisions and they go in to explain why they made those decisions they they do it with almost like a like an exaggerated sense of confidence like it was obvious you know, like it's totally obvious that you should be in in these kinds of markets. It's totally obvious you should use these kinds of fall controls. And and I think what it does, it puts investors, the end investors, the fund selectors in the position of basically saying, "All right, well, now I've got to challenge you and I've got to question you because if this is such a wonderful idea, why didn't you do it 5 years ago? Why did you do it 10 years ago?" You know, show me evidence that this incremental improvement that that you made, and I think that's just a narrative issue with the space. So I would love to hear from panelists in terms of when you've looked at some of these design decisions. You know, are there circumstances where you think it's really obvious and it's an absolute statistical no-brainer to make some sort of a modification or is it always as our experience has been kind of a judgment call at the end of the day? >> So I'm going to turn it over to you first, but actually I just want to maybe uh warn Rich a little bit that I actually would love to hear his thoughts as he is designing a lot of models. or uh and just to make sure that he's still awake down there in in Australia. But you off you go first. >> Yeah. I I think the the obviousness doesn't come from uh necessarily from better, but I think the obviousness comes from uh what are we providing to investors which may be different to what everybody else is is providing. And I think you've gone through this exercise with the ETF. It says I'm going to concentrate on the lowcost market and I'm going to concentrate on the most liquid futures which have the highest beta to what I call global macro factors. Okay. So in in your case, >> it's not necessarily that you know the exact timing, the exact period, the exact market really make a difference, but the the general principle is I'm going to try to harvest alpha from the highest from the highest financialized universe. Okay? And when you come for you know if if you look at at what my universe is where we try to find the market which have the least beta or they have a beta to a factor like inflation in the commodity universe which is really very different to this universe. What are what is one providing to the investor? You say I'm providing something different which is a risk factor which the sock gen index doesn't really provide you and most CTAs don't necessarily. And again you I think in the case of volatility driven uh trading trading strategies or maybe high frequency trend followers what am I providing? I'm trying to provide convexity right so so the idea is that if we look at what the investor what the investor wants to construct in for his portfolio or for her portfolio we are saying what are you actually holding at the moment and what are you holding CTAs for. Okay. And different investors will have a different criteria. What are they looking to build and depending on that it will be actually quite obvious what they actually want to hold. So you know many a times I would have a discussion with an investor and I say listen actually you don't really want alternative markets. It makes no sense to you. You're not buying it for like you know you don't need insurance against you know Japanese power uh price spike. That just completely doesn't help you. go for, you know, a really lowc cost solution, uh, long-term trend in equities. Is that kind of what you're looking to harvest? That's fine. Okay. So, obviousness is not when, you know, it's not a quant decision in terms of the implementation. We can argue until we are blue in the face whether volatility adjustment, you know, the style that that Mark spoke about, whether this style or that style are different. We obviously want different CTAs that we invest in to be solid, right? We want them to know what they're doing and that I think comes from talking to the to the people who are actually running those CTAs. But I think the obviousness comes not from the mathematics but it comes from what are the factors that I'm I'm I am looking to buy what exposures I'm looking for like you know I want inflation I want commodities I want I want equity growth I'm going to get to sort of like you know uh the ETF space um and that's that's I think where the obviousness come from. Now, Rich, I'm going to turn it over to you, but I will say I was reminded about our conversation about categorizing, for example, CTAs or trend followers into different categories. Um, just to give you a little bit of of uh inspiration to your uh to your thoughts. >> Yeah. Well, look, I I'll respond to Andrews first. um you know this this call on um whether design decisions are ever obvious and um I inevitably find when people say that it's from a position of hindsight um so it's very easy from that position to actually say something's very obvious but I find design decisions are never obvious and there's inevitably tradeoffs in everything I do in in in amended designs there's inevitably tradeoffs so you know I I tend to think that you know we dealing with uncertainty on the right hand side of the chart going forward. Um I you know if if we're relying on a a precedent regime I tend to think that's overfitting curve fitting. Um so you know I I I tend to sort of take the opposite view that um these design decisions are never obvious and um uh it it's only you know when we saw Winton for instance you know um I think it was in about uh uh you know that that decade long winter win were really questioning their allocation towards trend and they were significantly reducing their trend that was That was a a design decision effectively from from Winton's perspective which was accompanied by you know a really nasty regime that was a precursor to that and they probably paid the price of this behavioral drift um and then suddenly we found this resumption back of back into trend. So that that that that's how I'd view it. But look, in in relation to um um Nick's question about dispersion, I actually think it's very healthy um in in our trend following sector because I think that um I think where investors tend to get it wrong or the assumption is that we're a homogeneous clan. We're we're a pretty broad church and we have all of our uh intricacies, you know, our design intricacies. Some of us are outlaw hunters. Some of us are chasing a benchmark. Some of us are seeking crisis alpha. Some of us are are doing this, doing that. So, you know, I think it's this this dispersion is is healthy. Um, but I think it's upon ourselves to have to explain to the investors what are the objectives we're seeking with our approach. Unfortunately, in this world of trend following, we all tend to be treated by the same metrics, the same performance metrics. But, you know, um the metrics that uh might justify someone who wants a a smooth portfolio outcome is going to be different to the metrics that satisfy someone that rides the Brocking Bunko with with an outlier, for instance. Um, so there are different metrics for different objectives and I think it's it's really probably in the interests of the the trend following camp to start categorizing themselves into what are the objectives they're seeking because there's nothing more disconcerting to an investor when they think that they're going into the broad church of trend following to find that their expectations are totally uh dismayed when um that they suddenly find that they're they're trading a Malaney when they wanted smoothness, you Um, so I think it's up to us to be able to give further information to our investors to say what are the objectives we're seeking? What are the performance returns to expect from this outcome? Um, why did we perform badly during this period? Well, for our particular objectives, there were no outliers in this particular, you know, period of time. So, we're not we're not in the game to chase returns. We're there to address these outliers when they emerge. um under under uncertainty. So um that therefore changes the expectation of the investor to say oh well there were no outliers during this particular period. I totally understand that and that accounts for the poor performance as opposed to the expectation that um you know the broad church of trend following did well in this particular period. So why didn't you do well? Well well it's up to you to explain what your objectives were. >> Yeah. No, absolutely. Katie, over to you. Yeah, maybe I'm a good person to answer this question too, just given sort of that we do some replication and CTA like strategies and we do um more classic uh full asset trend following. I do agree that it's really about what is the objective of the program and it's about making the decisions and design decisions based on that objective. And I think there's very different objectives. For example, we believe that a CTA like product that does use liquid markets that gives you something similar to CTAs is a great one-stop solution for somebody who doesn't have any CTAs and doesn't want to make the selection choice. But at the same time, we work with institutions that have risk mitigation strategies that want to understand exactly how everything works and they understand why and when things do or don't work and the decisions are very different in those type of programs by design of the investor, what the investors are actually looking for to add to their portfolio. So I'd say we tend to not I wouldn't be a person that would say anything is obvious and in fact in finance like we all know that we're in the you know such a space where so many there are no obvious decisions but there are some decisions that are well thought through and I think by thinking about what the objective of your program is there are decisions that you will make to be consistent with that objective and I think that you have me mentioned this as well just this concept of a client that's looking for convexity is going to you're going to make a different decision than a client who's looking for CTA like returns. Um, and so I don't think any of those things are obvious. I just perhaps thinks in that in that audience when we were talking, Andrew, you definitely see sometimes a little bit of spiciness of a lot of conviction in what they do. And so maybe it comes across as as sounding obvious, but from my perspective, none of those things are ever obvious. >> Right. Mark, over to you, >> Rich. I'm going to take the exception to your idea that dispersion is healthy. I I think it actually is unhealthy in the sense is that then it's harder to classify what is a CTA. uh and in some senses that this has been an age-old problem that there's a difference between trend followers and someone who could be a CTA because they might have other strategies embedded with trend following. And so I think that this really puts the burden on managers to try to explain or provide a narrative to what they do that's unique and special and why they would do better in some environments and do worse in others. So I think that if anything is is that you go through other hedge fund strategies there probably is less dispersion and so it's a so it's a little bit easier to say there are factors that you could define that are associated with those strategies that we could actually then describe the strategy and then you know potentially replicate. Now, it's healthy in the senses that that that allows for a lot of different uh differentiation across managers, but I think it's unhealthy for investors because they have a harder time now trying to figure out who who is the right manager to fill into this bucket. >> Yeah, you are. >> Well, dispersion basically means diversification. Um, and that is not actually a bad thing for investors to have. So a lot of the time we would have a discussion with an investor and they would say well actually holding a CTA is actually very difficult because you know it's a sharp one strategy maybe a little bit less maybe a little bit more um and it's kind of difficult to to sit within my IC and they will come to us and say how come they've lost money this year okay so um I think what is nice about having dispersion within the CTA universe is that there are different sources of alpha like fast or slow, alternative markets, liquid futures, right? And that means that actually if you want to create a a joint portfolio which has a little bit of this, a little bit of that. Each of these things will give you alpha overall, I'm actually going to get something which is much easier for me to hold, which has a CTA like behavior but actually has a higher alpha. So I think this Persian within the CTA universe is actually very helpful. Um, of course, if you go and you say, well, I don't have a pure CTA here. I've got some other strategy inside, then of course all better off. But I think generally in terms of the discussions, the investors like to have, you know, here's seven macro traders and all of these macro traders are actually doing different things and investors like it and they don't find that like why why aren't you all getting the same returns, right? They can cope with that with that ambiguity, right? You never people you never hear you know different investors complaining why are all my long short strategies looking the same right because all long short strategies are actually looking very different um and that's an opportunity for them that's not a I don't think that's a barrier um I think it's it's it would be nice for the CTA industry to target different risk factors and in fact I think that having a homogeneous high correlation to sort of an index in some sense is is is a problem because they go and they meet different CTAs and they get exactly the same story and they get exactly sort of the the same sort of index-like behavior and then it's actually very difficult for them to tell each of the CTAs apart. Um, I think it's much nicer, I think, the way that Katie is talking about it. If different programs have different objectives and they target different risk factors, then it actually gives the investor a true choice and also clarity about when each of these component is likely to perform and that gives them much more confidence in allocate and just wait for that strategy to perform and do its job when it's supposed to be doing its job. >> Andrew, over to you. Yeah. So I look I think going back to the next point about about growing the pie and adoption. I think one of I like I honestly I think this is a challenge right outside of the I think this is a group of people who are very very technically sophisticated um I think who spend a lot of time in yo in terms of the clients you're talking to if they have a preference function around Japanese power markets like like then you're you're talking to the 0.01% um uh uh of the world out there. I think I think the challenge that that people are waiting to hear on the other side is is is the human element of like no one on this if we' done this four years ago would have predicted the outcome that that Nick laid out in terms of very very long-term models work you know like major markets work over the next three years. What most people are looking for is somebody to say, I knew that was going to happen and I reconfigured my models because of that, right? And and and I think that's the that's the challenge in this space is that when it goes wrong and and know when it goes wrong, I mean just like the markets don't behave as expected and you get bad luck and you get, you know, you do worse than expected or or or a strategy that that's been performing well does poorly or vice versa. Um I think that randomness is very very hard for most allocators because when dispersion feels to a typical person looking at the space that um that it just stuff happens and markets behave in unexpected ways, it's very very different than the narrative you get in equity long short. Like if I'm a if I'm an equity long short guy with a 30-year track record and I'm getting killed because but then I have a narrative around well look this looks a lot like this period of time when it came round bounding back and and look you know all those stocks that I was smart buying them at six times EBDA now because of this crazy markets all trading at four times EBDA there's always a compelling narrative a rebound narrative and I think I just my observation is the space really struggles with that um is and and a lot of the discussion in the space is about kind of it's it's to me this is a very very human business of people making very human decisions around how they're decide to build their models but but but the explanation is often of a technical one that I think most allocators it just leaves them feeling flat. >> Yeah. Before I get over to Alan and his topic uh I know that Nick had a a comment to this. Uh >> I'm actually glad that my question brought a lot of debate. Maybe I summarize it. I think we all agree that some dispersion is good, but too much of it is going to create too much fuss for us understanding what's going on, right? I think that's what I'm kind of trying to summarize out of that. Beyond that, I actually agree with some of the comments that Rich made on the on the question about um what can be objective in the design. I think it's just a matter of aligning your culture to avoid overfitting and address the blind spots and acknowledge the trade-offs and make a decision that is in line with what your clients are basically here to to solve for. So that that's all on my side. >> Well, Alan, all the way from Dublin, what's on your mind? >> Well, I feel like we've been skirting around my topic already quite a bit. um because it's back to kind of manager evaluation and how long you need. So we've kind of been touching about this dispersion theme and how performance can go up and down a lot and and then obviously what you tend to see from investors and investors I work with they'll say oh this is working in this environment or this manager is doing well and the kind of assumption is that it's going to continue. Um and then you talk to quants or statisticians and they'll say actually the amount of data you need to evaluate a manager and get to get conviction that they're realized sharp is or true sharp is way longer than what most investors realize that we could be talking 10 years plus. So on the one hand you have you know as Andrew was talking about this aversion to randomness or misappreciation of man randomness and you've got whenever whenever anybody's doing well there's a narrative to support that. Uh and then there's the bias the recency bias and then you've got other biases career risk very hard to defend you know an underperforming manager. So I mean there's a couple of questions. One is for the for the statist statisticians for the quants just it it is useful to to reiterate how long how much data do we actually need to make that manager uh assessment and then practically I mean as as kind of educators in the space what should we be saying to investors? I don't think we say this enough that we can't we really rely on performance data. We kind of say it but then kind of default back to hey look at the performance. As as as educators in the space, what should we be saying to investors to kind of shift their mindset away from just making decisions based on one, two, even three years performance? >> Rob, perfect that you raised your hand. >> Yeah, I'm going to do the maths. That's okay because that's the easy one. Much harder to know how to talk to clients about uncertainty and randomness and unfortunately I don't have to. Um yeah, so um the two things that kind of input into the equation as to how long you need are the the degree of outperformance. So if someone's outperforming by they say say someone's outerforming the S&P by 20% a year, then it's not going to take you very long to conclude that statistically they are better than the S&P 500. Um so that's the thing to be clear. Are we talking about what a statistician would call a test against a hypothesis of zero returns? Other words, can we be confident this return or additional shown is positive or are we talking about a specific number? So, we looking for an out performance by a given benchmark. Obviously, it's much harder to say this person's 20% better than to just say they're better. Um, and the other thing that comes into it, and this kind of feeds back into the conversation already had, is correlation. So, if if two variables are highly correlated, then you need much less data to determine if one has got a higher return than the other one. Um, so the the more weird and obscure your CTA is versus everybody else, the harder it will be to say that they're doing a better job. Whereas if your mandate, and I'm not pointing at anything here, but if your mandate's to beat say um, you know, a CTA index by 50 basis points a year, um, and to get a very high correlation to that index in the process, then okay, 50 basis points isn't a big outperformance. So that will take a while. But the fact that you're trying to get a very higher correlation against something means that it will you will be able to find that kind of out performance quicker. But you know in general terms we are talking about years if not decades often to to actually say with something like a 5% critical value which is kind of a you know what stat editions use. To be honest 5% is ridiculous. Like most most people would be happy if they were 75% confident that X was better than Y. So, you know, we can probably use much shorter periods of time than you would for a pure statistical significance. Um, but yeah, I think the more the more dispersion is in the industry, the harder it is to say whether people are outperforming other people. >> Um, because it's just random really. >> Yeah, you have. >> Um, I don't think the decision is actually living in the space of statistics to be honest. Uh, because by the time by the time you think you actually have a statistical point of point of view, uh, then the universe have changed, right? So in just to give you an example you're sitting in the alternative markets people it it took about you know 10 years for people to think oh my god alternative markets are great and then like you know it takes a year or two of underperformance and then suddenly oh it's over capacity the place is too crowded um the universe has changed in some sense okay so um I think what you're looking for um if in determining a manager is really um understanding what the manager is and understanding that they are solid, that they're doing things that are sensible. I think Katie talked about um understanding that we have wellthoughtout design questions. They don't have to be the best, but they have to be solid. They have to be understood. And you kind of need to understand also how the manager would um you know would have proper control over trading, over data, over like you need to have a confidence that the person actually knows what he's doing. and the way that he would response to crisis or she would respondse to crisis. So I think one of the interesting questions um that we have seen from investors in the last three years was about how do you respond to the you know the current crisis in in your CTA universe while being true to the CTI hypothesis. You know what aspects of your trading you can actually improve on. Is it execution? Is it risk management that you can do things while actually being committed to trend following? And I think um Winon was a great example where you go through a situation where a manager basically drifts away from the core competency uh and then has to retrace the steps and and drift back into into CTA land. So I think the decisions to to which manager you want is really about actually how uh how well thought out his thought process is. How solid are the you know execution data ingestion computation release process you know operational due diligence and then all the way back to all the way to execution. Um and then how they handle the changing dynamics in the market I think is is another thing that uh I think some some managers behave differently to others. >> All right, Alan, I hope you got some uh thoughts on that. Nick, actually you just wanted to jump in here. That's great. >> Maybe I had a few comments. I mean slightly different perspective obviously looking into the QIS space know we launch investment strategies we kind of monitor them sometimes we have to update them and so on and so forth I guess if I were to paraphrase the question from Alan when when do we change the strategy or you know what is the point beyond which we feel that a particular strategy is not viable enough and I don't know the market has changed I would say and I agree with you a lot it's extremely hard to come up with a statistical framework that gets you there in reality If there's a fundamental reason why we build a systematic strategy, we have to assess the reasons why the initial premise is not there yet. Maybe the markets became less liquid or more expensive and that's a different story. Um, so if it's not a cost discussion, it has to be the reasons why the premium was there in the first place. I'll give you a good example. Now, commodity congestion, some of you might be very much well aware of this premium. It's all about pre-rolling those uh benchmark roles in those commodity futures because you'd expect a lot of money flowing into kind of passive commodity exposure. So if you were to pre-roll and effectively buy whatever or less we'll be buying later and selling whatever else we'll be selling later, you will be there on a long short basis with a bit of a premium. That's a liquidity premium. So there came a time around 2018 2019 that we started looking into this performance kind of flattening out. Now obviously the question here is you know where is the premium and obviously spending a bit amount of time you realize that you know there was not too much um of commodity allocation or maybe investors became a bit more sophisticated they would end up kind of rolling on the back end of the curve that was not a crowding situation. You know the subtle difference here is that you know crowding would make the strategy underperform not flat out because flattening out um you know a an excess supply or an excess demand for a particular asset that is beyond any risk takingaking should just flatt it out right and and you know maybe I can now use one of the points that J made in the beginning. You know if we see all this demand about you know auto calls or maybe structure products that have specific impact on on on the supply and the demand for volatility maybe dispersion becomes a better theme but there is no particular reason to actually feel that tomorrow you know investors become risk loving and we're not over buying those index boots. So when can we come out come out and say look you know selling volatility is not a profitable strategy it's actually quite impossible because that goes against the economic nature of how investors operate. So I think bottom line the bar is extremely high. Of course we can change things but that's more procedural and more maybe methodological but there are a few cases whereby if the fundamental reason underlying a strategy is more of a structural temporary shift then there could be a case that we know we start considering how the strategy may be um should be redesigned or maybe removed right that's not about manager evaluation but I think it kind of touches upon that from a pure kind of design perspective. So that's that's basically my take on that. >> Sure. Sure. >> Yeah. I think QIS brings out um another aspect which is um different allocators prefer alignment. So one of the FM issues in in in CTAs is we've seen a drift from performance only from performance related uh pay to um you know flat management fee. Um and and that's great. So, you know, in a lot of the spaces in the in the in the universe of the ETF, um, a lot of a lot of investors want that and that's a select that's part of the selection process is, you know, it's just it's just cheaper. Um, equally, you might have a situation where managers where allocators actually want you to rely. They want you to have skin in the game. They want you to to know that um, how you are making money out of that product and they want you to have the same exposures to them. Um, and that's very different from whether you go into sort of the ETF, the QIS, the sort of the management fee only CTAs all the way to of the more the more niche CTAs. Um, so again, it's a choice for you in terms of your manager selection. Uh, but we find that actually is important to the people who invest in us. >> On that note, let's wrap up part one of our year- end group conversation. We hope that you've enjoyed it as much as we did making it for you. And if you want to show your appreciation for all the work that the amazing co-hosts put into making these episodes each week, I would encourage you to head over to Apple Podcast or Spotify or wherever you listen to your podcast and leave a nice rating and review. We really do appreciate all of them. Next week, we will publish the second part of our group conversation. So, we hope, of course, that you'll be back for more and to hear our 2026 outrageous predictions. From all of us at Top Traders Unplucked, thank you so much for listening. We look forward to being back with you next week. Until next time, happy holidays and take care of yourself and take care of each other. 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 in 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.